U.S. patent application number 11/342337 was filed with the patent office on 2007-06-07 for method and apparatus for spectral-beam combining of high-power fiber lasers.
Invention is credited to Andrew J. W. Brown, Charles E. Hamilton, Eric C. Honea, Charles A. Lemaire, Anping Liu, Thomas H. Loftus, Roy D. Mead.
Application Number | 20070127123 11/342337 |
Document ID | / |
Family ID | 38118450 |
Filed Date | 2007-06-07 |
United States Patent
Application |
20070127123 |
Kind Code |
A1 |
Brown; Andrew J. W. ; et
al. |
June 7, 2007 |
METHOD AND APPARATUS FOR SPECTRAL-BEAM COMBINING OF HIGH-POWER
FIBER LASERS
Abstract
Apparatus and method for spectral-beam combining light from a
plurality of high-power fiber lasers that, in some embodiments, use
two substantially identical diffraction gratings in a parallel,
mutually compensating configuration to combine a plurality of
separate parallel input beams each having a slightly different
successively higher wavelength into a single output beam of high
quality. In other embodiments, a single diffraction grating is used
to combine a plurality of different wavelengths, wherein the input
laser beams are obtained from very narrow linewidth sources to
reduce chromatic dispersion. In some embodiments, diagnostics and
adjustments of wavelengths and/or positions and angles are made
dynamically in real time to maintain the combination of the
plurality input beams into a single high-quality output beam.
Inventors: |
Brown; Andrew J. W.; (Brier,
WA) ; Honea; Eric C.; (Seattle, WA) ; Loftus;
Thomas H.; (Seattle, WA) ; Mead; Roy D.;
(Edmonds, WA) ; Hamilton; Charles E.; (Kenmore,
WA) ; Liu; Anping; (Big Flats, NY) ; Lemaire;
Charles A.; (Apple Valley, MN) |
Correspondence
Address: |
LEMAIRE PATENT LAW FIRM, P.L.L.C.
PO BOX 11358
ST PAUL
MN
55111
US
|
Family ID: |
38118450 |
Appl. No.: |
11/342337 |
Filed: |
January 26, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60647747 |
Jan 26, 2005 |
|
|
|
60703824 |
Jul 29, 2005 |
|
|
|
Current U.S.
Class: |
359/556 |
Current CPC
Class: |
G02B 27/1086 20130101;
G02B 27/123 20130101; G02B 27/1006 20130101; G02B 27/108 20130101;
G02B 27/145 20130101; G02B 27/144 20130101 |
Class at
Publication: |
359/556 |
International
Class: |
G02B 27/64 20060101
G02B027/64 |
Goverment Interests
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] This work was supported by the U.S. Air Force under contract
F29601-98-D-0190 and by the U.S. Navy under contract
N00178-04-C-3045. The U.S. Government has a paid-up license in this
invention and the right in limited circumstances to require the
patent owner to license others on reasonable terms as provided for
by the terms of these contracts.
Claims
1. A method comprising: providing a plurality of fiber lasers that
generate laser beams including a first fiber that generates a first
laser beam and a second fiber that generates a second laser beam;
spectrally combining the plurality of laser beams into a single
output beam; wavelength tuning the first fiber to generate the
first laser beam at a first wavelength; wavelength tuning the
second fiber to generate the second laser beam at a second
wavelength; detecting that one of the laser beams has become
misaligned relative to the single combined beam; determining that
the first laser beam is the misaligned one; and adjusting the
wavelength tuning of the first fiber in order that the first laser
beam is aligned relative to the single combined beam.
2. The method of claim 1, wherein the spectrally combining further
includes: providing a first diffractive element; introducing a
first chromatic dispersion into the first laser beam with the first
diffractive element; introducing a second chromatic dispersion into
the second laser beam with the first diffractive element; and
providing a second diffractive element; and introducing a third
chromatic dispersion into the first laser beam with the second
diffractive element, wherein the third chromatic dispersion is a
compensating dispersion that negates at least a portion of the
first chromatic dispersion from the first laser beam; and
introducing a fourth chromatic dispersion into the second laser
beam with the second diffractive element, wherein the fourth
chromatic dispersion is a compensating dispersion that negates at
least a portion of the second chromatic dispersion from the second
laser beam.
3. The method of claim 2, wherein the providing of the first
diffractive element and the second diffractive element includes
providing dielectric diffractive gratings having the same
diffractive pattern on both the first diffractive element and the
second diffractive element.
4. The method of claim 2, further comprising positioning of the
second diffractive element so a diffractive surface of the second
diffractive element is parallel to a corresponding diffractive
surface of the first diffractive element.
5. A method comprising: providing a plurality of fiber lasers that
generate laser beams including a first fiber that generates a first
laser beam and a second fiber that generates a second laser beam;
providing a plurality of Yb-doped multiply clad optical fibers
including a first fiber and a second fiber; pumping an inner
cladding of each of the plurality of fibers with pump light from
one or more laser diodes; amplifying, with the first fiber, the
first laser beam at a first wavelength; amplifying, with the second
fiber, the second laser beam at a second wavelength; spectrally
combining the plurality of laser beams into a single output beam;
wavelength tuning the first fiber to generate the first laser beam
at the first wavelength; wavelength tuning the second fiber to
generate the second laser beam at the second wavelength; detecting
that one of the laser beams has become misaligned relative to the
single combined beam; determining that the first laser beam is the
misaligned one; and adjusting the wavelength tuning of the first
fiber in order that the first laser beam is aligned relative to the
single combined beam.
6. The method of claim 5, further comprising: filtering the first
laser beam to a full-width half-maximum linewidth of about one
nanometer or less; filtering the second laser beam to a full-width
half-maximum linewidth of about one nanometer or less; pulsing the
first laser beam to a pulse length of about ten nanoseconds or
less, and sufficiently short to substantially prevent stimulated
Brillouin scattering (SBS) buildup in the amplifying of the first
laser beam; and pulsing the second laser beam to a pulse length of
about ten nanoseconds or less, and sufficiently short to
substantially prevent SBS buildup in the amplifying of the second
laser beam.
7. The method of claim 6, further comprising: detecting that one of
the laser beams has become angularly misaligned relative to the
single combined beam; determining which laser beam is the angularly
misaligned one; and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam, wherein the detecting of the angularly misaligned beam and
the determining of which laser beam is the angularly misaligned one
are based on a timing of a pulse of one of the laser beams.
8. The method of claim 5, further comprising: detecting that one of
the laser beams has become angularly misaligned relative to the
single combined beam; determining which laser beam is the angularly
misaligned one; and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam.
9. The method of claim 8, wherein the detecting of the angularly
misaligned beam and the determining of which laser beam is the
angularly misaligned one are performed while one or more of the
other beams are active.
10. The method of claim 9, wherein the determining of which laser
beam is the angularly misaligned one is performed during a time
when a plurality of the other laser beams are on, and includes:
changing a power value of a first laser beam; detecting a
corresponding change in a misaligned beam; and changing the angle
of the first laser based on the detecting of the corresponding
change in the misaligned beam.
11. A method comprising: providing a plurality of laser beams
including a first laser beam and a second laser beam; wavelength
tuning a first fiber to generate the first laser beam at a first
wavelength and having a linewidth of about 1 nm or less; wavelength
tuning a second fiber to generate the second laser beam at a second
wavelength and having a linewidth of about 1 nm or less; and
spectrally combining the plurality of laser beams into a single
output beam having an output power/area of about 10 W/cm.sup.2 or
more using one or more high-efficiency dielectric diffractive
gratings.
12. The method of claim 11, further comprising: temporally forming
the first laser beam into a first serial plurality of pulses each
having a pulse length of about 10 ns or less; and temporally
forming the second laser beam into a second serial plurality of
pulses each having a pulse length of about 10 ns or less.
13. A method comprising: providing a plurality of laser beams
including a first laser beam and a second laser beam; wavelength
tuning a first fiber to generate the first laser beam at a first
wavelength and having a linewidth of about 1 nm or less; wavelength
tuning a second fiber to generate the second laser beam at a second
wavelength and having a linewidth of about 1 nm or less; spectrally
combining the plurality of laser beams into a single output beam
having an output power/area of about 10 W/cm.sup.2 or more using
one or more high-efficiency dielectric diffractive gratings,
temporally forming the first laser beam into a first serial
plurality of pulses each having a pulse length of about 10 ns or
less; and temporally forming the second laser beam into a second
serial plurality of pulses each having a pulse length of about 10
ns or less, wherein pulses of the first serial plurality of pulses
are alternated with pulses of the second serial plurality of
pulses.
14. The method of claim 13, further comprising: detecting that one
of the laser beams has become misaligned relative to the single
combined beam; determining that the first laser beam is the
misaligned one; and adjusting the wavelength tuning of the first
fiber in order that the first laser beam is aligned relative to the
single combined beam.
15. The method of claim 11, wherein the spectrally combining
further includes: providing a first diffractive element;
introducing a first chromatic dispersion into the first laser beam
with the first diffractive element; providing a second diffractive
element; introducing a second chromatic dispersion into the second
laser beam with the second diffractive element; and providing a
third diffractive element; and spectrally combining the first and
second laser beams and introducing a third chromatic dispersion
into the first laser beam with the third diffractive element,
wherein the third chromatic dispersion is a compensating dispersion
that negates at least a portion of the first chromatic dispersion
from the first laser beam, and introducing a fourth chromatic
dispersion into the second laser beam with the second diffractive
element, wherein the fourth chromatic dispersion is a compensating
dispersion that negates at least a portion of the second chromatic
dispersion from the second laser beam.
16. The method of claim 15, wherein the providing of the first
diffractive element and the second diffractive element includes
providing dielectric diffractive gratings having the same
diffractive pattern on both the first diffractive element and the
second diffractive element, and positioning the second diffractive
element so a diffractive surface of the second diffractive element
is approached by the laser beams at an angle corresponding to an
angle the beams left the first diffractive element.
17. A method comprising: providing a plurality of laser beams
including a first laser beam and a second laser beam; wavelength
tuning a first fiber to generate the first laser beam at a first
wavelength and having a linewidth of about 1 nm or less; wavelength
tuning a second fiber to generate the second laser beam at a second
wavelength having a linewidth of about 1 nm or less; spectrally
combining the plurality of laser beams into a single output beam
having an output power/area of about 10 W/cm.sup.2 or more using
one or more high-efficiency dielectric diffractive gratings,
providing a plurality of Yb-doped large-mode-area (LMA)
optical-amplification fibers each operating substantially in the
fundamental mode, including a first fiber and a second fiber;
pumping of each of the plurality of fibers with pump light from one
or more laser diodes; amplifying, with the first fiber, the first
laser beam at a first wavelength; amplifying, with the second
fiber, the second laser beam at a second wavelength; filtering the
first laser beam to a full-width half-maximum linewidth of about
one nanometer or less; filtering the second laser beam to a
full-width half-maximum linewidth of about one nanometer or less;
pulsing the first laser beam to a pulse length of about ten
nanoseconds or less, and sufficiently short to substantially
prevent SBS buildup in the amplifying of the first laser beam; and
pulsing the second laser beam to a pulse length of about ten
nanoseconds or less, and sufficiently short to substantially
prevent SBS buildup in the amplifying of the second laser beam.
18. The method of claim 17, further comprising: detecting that one
of the laser beams has become angularly misaligned relative to the
single combined beam; determining which laser beam is the angularly
misaligned one; and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam, wherein the detecting of the angularly misaligned beam and
the determining of which laser beam is the angularly misaligned one
are based on a timing of a pulse of one of the laser beams.
19. The method of claim 18, wherein the detecting of the angularly
misaligned beam and the determining of which laser beam is the
angularly misaligned one are performed while one or more of the
other beams are active.
20. An apparatus comprising: a plurality of fiber lasers that
generate laser beams including a first fiber that generates a first
laser beam and a second fiber that generates a second laser beam;
means for spectrally combining the plurality of laser beams into a
single output beam; means for wavelength tuning the first fiber to
generate the first laser beam at the first wavelength; means for
wavelength tuning the second fiber to generate the second laser
beam at the second wavelength; means for detecting that one of the
laser beams has become misaligned relative to the single combined
beam; means for determining that the first laser beam is the
misaligned one; and means for adjusting the wavelength tuning of
the first fiber in order that the first laser beam is aligned
relative to the single combined beam.
21. The method of claim 5, further comprising: filtering the first
laser beam to a linewidth sufficiently narrow and amplifying the
first laser beam to a sufficiently high extent to otherwise cause
stimulated Brillouin scattering (SBS) buildup in the amplifying of
the first laser beam; filtering the second laser beam to a
linewidth sufficiently narrow and amplifying the firstlaser beam to
a sufficiently high extent to otherwise cause SBS buildup in the
amplifying of thesecond laser beam; pulsing the first laser beam to
a pulse length sufficiently short to substantially prevent SBS
buildup in the amplifying of the first laser beam; and pulsing the
second laser beam to a pulse length sufficiently short to
substantially prevent SBS buildup in the amplifying of the second
laser beam.
22. The method of claim 21, further comprising: detecting that one
of the laser beams has become angularly misaligned relative to the
single combined beam; determining which laser beam is the angularly
misaligned one; and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam, wherein the detecting of the angularly misaligned beam and
the determining of which laser beam is the angularly misaligned one
are based on a timing of a pulse of one of the laser beams.
23. A method comprising: providing a plurality of laser beams
including a first laser beam and a second laser beam; wavelength
tuning a first fiber to generate the first laser beam at a first
wavelength and at a sufficiently narrow linewidth and amplifying
the first laser beam to a sufficiently high extent to otherwise
cause stimulated Brillouin scattering (SBS) buildup in the
amplifying of the first laser beam but temporally forming the first
laser beam into a first serial plurality of pulses each having a
pulse length no more than about 10 nanoseconds, in order to
substantially prevent SBS buildup in the amplifying of the first
laser beam; wavelength tuning a second fiber to generate the first
laser beam at a second wavelength and at a sufficiently narrow
linewidth and amplifying the second laser beam to a sufficiently
high extent to otherwise cause SBS buildup in the amplifying of the
second laser beam but temporally forming the second laser beam into
a second serial plurality of pulses each having a pulse length no
more than about 10 nanoseconds, in order to substantially prevent
SBS buildup in the amplifying of the second laser beam; and
spectrally combining the plurality of laser beams into a single
output beam using one or more high-efficiency dielectric
diffractive gratings.
24. A method comprising: providing a plurality of laser beams
including a first laser beam and a second laser beam; wavelength
tuning a first fiber to generate the first laser beam at a first
wavelength and at a sufficiently narrow linewidth and amplifying
the first laser beam to a sufficiently high extent to otherwise
cause stimulated Brillouin scattering (SBS) buildup in the
amplifying of the first laser beam but temporally forming the first
laser beam into a first serial plurality of pulses each having a
pulse length sufficiently short to substantially prevent SBS
buildup in the amplifying of the first laser beam; wavelength
tuning a second fiber to generate the first laser beam at a second
wavelength and at a sufficiently narrow linewidth and amplifying
the second laser beam to a sufficiently high extent to otherwise
cause SBS buildup in the amplifying of the second laser beam but
temporally forming the second laser beam into a second serial
plurality of pulses each having a pulse length sufficiently short
to substantially prevent SBS buildup in the amplifying of the
second laser beam; spectrally combining the plurality of laser
beams into a single output beam using one or more high-efficiency
dielectric diffractive gratings; detecting that one of the laser
beams has become misaligned relative to the single combined beam;
determining that the first laser beam is the misaligned one; and
adjusting the wavelength tuning of the first fiber in order that
the first laser beam is aligned relative to the single combined
beam.
25. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 9 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 9 nanoseconds.
26. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 8 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 8 nanoseconds.
27. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 7 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 7 nanoseconds.
28. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 6 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 6 nanoseconds.
29. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 5 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 5 nanoseconds.
30. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 4 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 4 nanoseconds.
31. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 3 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 3 nanoseconds.
32. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 2 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 2 nanoseconds.
33. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length no more than about 1 nanoseconds; and the temporally forming
of the second laser beam comprises temporally forming the second
laser beam into the second serial plurality of pulses each having a
pulse length no more than about 1 nanoseconds.
34. The method of claim 23, wherein: the temporally forming of the
first laser beam comprises temporally forming the first laser beam
into the first serial plurality of pulses each having a pulse
length of about 5 nanoseconds and a between-pulse off time of about
95 nanoseconds; and the temporally forming of the second laser beam
comprises temporally forming the second laser beam into the second
serial plurality of pulses each having a pulse length of about 5
nanoseconds and a between-pulse off time of about 95 nanoseconds.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This invention claims benefit of U.S. Provisional Patent
Application 60/647,747 filed on Jan. 26, 2005, titled "SPECTRAL
BEAM COMBINING OF HIGH POWER FIBER LASERS", and U.S. Provisional
Patent Application 60/703,824 filed on Jul. 29, 2005, titled
"PERIODIC FIBER TO SUPPRESS NONLINEAR EFFECTS IN RARE-EARTH-DOPED
FIBER AMPLIFIERS AND LASERS" which are each hereby incorporated by
reference in entirety.
FIELD OF THE INVENTION
[0003] The invention relates generally to high-power optical
amplifiers and lasers and more particularly to methods and
apparatus for combining the outputs of a plurality of optical
fibers into a single beam of excellent beam quality, as measured by
beam-waist size and dispersion angle.
BACKGROUND OF THE INVENTION
[0004] The broad gain bandwidth of conventional fiber-laser systems
allows for operation over a wide range of wavelengths, or even
tunable operation. For the simplest fiber laser system with cavity
mirrors having reflectivity across a broad range of wavelengths,
the output wavelength can be very broad and can vary with pump
power, fiber length, and/or other parameters. The power that can be
generated from fiber lasers and fiber-laser amplifiers can often be
limited by nonlinear optical effects in the gain and/or delivery
fibers used in the system.
[0005] It is desirable to produce high peak and average powers from
fiber lasers and amplifiers. Stimulated Brillouin scattering (SBS)
and other nonlinear effects such as self-phase modulation (SPM),
four-wave mixing (FWM), and stimulated Raman scattering (SRS) are
the main effects limiting the output power and pulse energy of a
fiber amplifier or laser. To suppress these effects in a fiber
amplifier/laser, it is desirable to use a rare-earth-doped
(RE-doped) fiber with a large core. The large core provides two
benefits: Spreading the light over a larger core decreases the
intensity driving the nonlinear processes, and increasing the
core/cladding diameter ratio increases pump absorption, enabling
the shortening of the fiber to further reduce nonlinearities. When
good beam quality is required, however, increasing the core
diameter of the fiber requires that the fiber numerical aperture
(NA) be decreased, in order that higher-order modes cannot
propagate in the fiber. Using relatively large-core, low-NA fibers
with mode-filtering techniques has been demonstrated to achieve
good beam quality, but there are practical disadvantages to the use
of such fibers. Fibers with very low values of NA exhibit large
bending losses, even for relatively large-radius bends. With fibers
having the lowest NA, the fiber must be kept quite straight,
otherwise the optical amplifier and/or laser has very low
efficiency as the bending loss becomes too high. Since a typical
laser oscillator or amplifier might require on the order of a meter
or more of gain fiber, the inability to coil the fiber has
precluded compact packaging of the fiber-laser system.
[0006] Stimulated Brillouin scattering (SBS) is a well-known
phenomenon that can lead to power limitations or even the
destruction of a high-power fiber-laser system due to sporadic or
unstable feedback, self-lasing, pulse compression and/or signal
amplification.
[0007] Even when a fiber amplifier or fiber laser is designed to
compensate for the above effects, there will be a limit on the
maximum power that can be obtained from a single fiber when scaling
to larger fiber sizes and/or lengths, pump powers, and the
like.
[0008] U.S. Pat. No. 6,192,062 to Sanchez-Rubio et al. entitled
"Beam combining of diode laser array elements for high brightness
and power" and U.S. Pat. No. 6,208,679 to Sanchez-Rubio et al.
entitled "High-power multi-wavelength external cavity laser
describe the fundamental techniques of spectral beam combining, and
both are incorporated herein by reference.
[0009] In some embodiments, the gratings used for spectral-beam
combining are "blazed," i.e., formed with V-grooves having sidewall
angles that are asymmetrical with respect to a vector normal to the
overall surface of the grating. U.S. Pat. No. 3,728,117 to
Heidenhain et al. entitled Optical Diffraction Grid" (incorporated
herein by reference) describes one method for making blazed
gratings having asymmetric grooves. U.S. Pat. No. 3,728,117 to
Heidenhain et al. entitled "Optical Diffraction Grid" (incorporated
herein by reference) describes a method for making blazed gratings
having asymmetric grooves. U.S. Pat. No. 4,895,790 to Swanson et
al. entitled "High-efficiency, multilevel, diffractive optical
elements" (incorporated herein by reference) describes a method for
making blazed gratings having asymmetric grooves using binary
photolithography to create stepped profiles. U.S. Pat. No.
6,097,863, titled "Diffraction Grating with Reduced Polarization
Sensitivity" issued Aug. 1, 2000 to Chowdhury (incorporated herein
by reference) describes a reflective diffraction grating with
reduced polarization sensitivity for dispersing the signals. The
Chowdhury grating includes facets that are oriented for reducing
efficiency variations within a transmission bandwidth and that are
shaped for reducing differences between the diffraction
efficiencies in two orthogonal directions of differentiation. U.S.
Pat. No. 4,313,648 entitled "Patterned Multi-Layer Structure and
Manufacturing Method" issued Feb. 2, 1982 to Yano et al.
(incorporated herein by reference) describes a manufacturing method
for a patterned (striped) multi-layer article.
[0010] U.S. Pat. No. 6,822,796 to Takada et al. entitled
"Diffractive optical element" (incorporated herein by reference)
describes a method for making blazed gratings having asymmetric
grooves with dielectric coatings. U.S. Pat. No. 6,958,859 to Hoose
et al. entitled "Grating device with high diffraction efficiency"
(incorporated herein by reference) describes a method for making
blazed gratings having dielectric coatings.
[0011] U.S. Pat. No. 5,907,436 entitled "Multilayer dielectric
diffraction gratings" issued May 25, 1999 to Perry et al., and is
incorporated herein by reference. This patent describes the design
and fabrication of dielectric grating structures with high
diffraction efficiency. The gratings have a multilayer structure of
alternating index dielectric materials, with a grating structure on
top of the multilayer, and obtain a diffraction grating of
adjustable efficiency, and variable optical bandwidth.
[0012] U.S. Pat. No. 6,212,310 entitled "High power fiber gain
media system achieved through power scaling via multiplexing"
issued 3Apr. 2001 to Waarts et al., and is incorporated herein by
reference. This patent describes certain methods of power scaling
by multiplexing multiple fiber gain sources with different
wavelengths, pulsing or polarization modes of operation is achieved
through multiplex combining of the multiple fiber gain sources to
provide high power outputs, such as ranging from tens of watts to
hundreds of watts, provided on a single mode or multimode fiber.
One method described by Waarts et al. is similar to that shown in
the present invention shown in FIG. 2A, described below, where a
plurality of input laser beams of differing wavelengths are
directed at different angles to a diffraction grating, which
diffracts the beams into a single output beam, however, this output
beam necessarily has a wavelength linewidth-dependent chromatic
divergence introduced by the grating. The present invention
includes many distinguishing features not in Waarts et al.
[0013] There is a need for improved laser systems, particularly
fiber lasers and/or fiber optical amplifiers, wherein the optical
outputs from a plurality of fibers and/or other lasers are combined
into a single beam.
BRIEF SUMMARY OF THE INVENTION
[0014] In some embodiments, the invention provides methods and
apparatus for spectral-beam combining the optical output from a
plurality of high-power fiber lasers in a manner that provides
improved or superior output beam quality.
[0015] In some embodiments, the present invention uses two parallel
gratings to combine the optical outputs from a plurality of optical
fibers, each having a successively higher, slightly different peak
wavelength. In some embodiments, the output beams from the
plurality of fibers are focussed into a row of parallel collimated
input beams that impinge on the first grating of the pair and are
each diffracted at a plurality of successively higher slightly
different angles that all impinge on the second grating of the pair
at a common overlapped area, whereupon they all diffract into a
single combined output beam having higher power and higher quality
(e.g., a small waist and small divergence), as compared to
conventional devices.
[0016] In some embodiments, two substantially identical diffraction
gratings are used in a parallel, mutually compensating
configuration, where each beam is first diffracted by the first
grating at wavelength-dependent angles to combine a plurality of
separate parallel input beams each having a slightly different
successively higher wavelength into an area on the second grating,
which diffracts the combined beams into a single output beam, while
simultaneously introducing a compensating dispersion that removes
the angular dispersion within each beam that was introduced by the
first grating. That is, the chromatic dispersion introduced to each
individual beam by the first grating is needed in order to have the
different-wavelength beams combine into a single beam, but the
chromatic angular dispersion introduced within each individual beam
is then removed by the second grating. In other embodiments, a
single diffraction grating is used to combine a plurality of
different wavelengths, wherein the input laser beams are obtained
from very narrow linewidth sources to reduce chromatic angular
dispersion (also called herein simply "chromatic dispersion"). In
some embodiments, these very narrow linewidth sources are pulsed
(e.g., 5-ns pulses, in some embodiments) to prevent SBS waves from
building up. In some embodiments, diagnostics and adjustments of
wavelengths and/or positions and angles are made dynamically in
real time to maintain the combination of the plurality input beams
into a single high-quality output beam.
[0017] Some embodiments of the present invention provide a method
that includes providing a first laser beam and a second laser beam;
introducing a first chromatic angular dispersion into the first
laser beam; introducing a second chromatic angular dispersion into
the second laser beam; introducing a third chromatic angular
dispersion into the first laser beam, wherein the third chromatic
angular dispersion compensates for the first chromatic angular
dispersion; introducing a fourth chromatic angular dispersion into
the second laser beam, wherein the fourth chromatic angular
dispersion compensates for the second chromatic angular dispersion;
and combining the first and second laser beams into a single output
beam.
[0018] In some embodiments, the introducing of the first chromatic
angular dispersion includes diffracting the first laser beam with a
first diffractive element, and the introducing of the second
chromatic angular dispersion includes diffracting the second laser
beam with the first diffractive element, and the introducing of the
third chromatic angular dispersion includes diffracting the first
laser beam with a second diffractive element, and the introducing
of the fourth chromatic angular dispersion includes diffracting the
second laser beam with the second diffractive element, introducing
the compensating chromatic angular dispersions to each respective
one of the plurality of diffracted laser beams. In some
embodiments, the first diffractive element and the second
diffractive element have the same diffractive pattern. In some
embodiments, the second diffractive element is positioned so a
diffractive surface of the second diffractive element is parallel
to a corresponding diffractive surface of the first diffractive
element.
[0019] Some embodiments of the present invention provide an
apparatus that includes a first diffractive element and a second
diffractive element, and a source of a plurality of light beams
directed to a plurality of locations on the first diffractive
element, wherein the second diffractive element is positioned
relative to the first diffractive element such that the plurality
of light beams diffracted from the plurality of locations on the
first diffractive element are directed to a single location on the
second diffractive element and are diffracted by the second element
into a single combined beam.
[0020] Some embodiments of the present invention provide a method
that includes providing a plurality of laser beams including a
first laser beam and a second laser beam, spectrally combining the
plurality of laser beams into a single output beam, wavelength
tuning the first fiber to generate the first laser beam at the
first wavelength, wavelength tuning the second fiber to generate
the second laser beam at the second wavelength, detecting that one
of the laser beams has become misaligned relative to the single
combined beam, determining that the first laser beam is the
misaligned one, and adjusting the wavelength tuning of the first
fiber in order that the first laser beam is aligned relative to the
single combined beam.
[0021] Some embodiments provide an apparatus that includes an
output diffractive element, and a source of a plurality of
substantially monochromatic light beams directed from different
angles to a single location on the output diffractive element,
wherein the output diffractive element spectrally combines the
plurality of light beams into a single beam, and wherein the
plurality of light beams includes a first light beam having a first
central wavelength and a second light beam having a second central
wavelength, a first adjustment apparatus operatively coupled to set
an adjustable characteristic of the first light beam, a second
adjustment apparatus operatively coupled to set an adjustable
characteristic of the second light beam, a detector operatively
coupled to detect whether one of the light beams has become
misaligned relative to the single combined beam, a diagnoser
operatively coupled to determine whether the first light beam is
the misaligned one and if so, to control the first adjustment
apparatus to adjust the adjustable characteristic of the first
light beam in order that the first light beam becomes aligned
relative to the single combined beam.
BRIEF DESCRIPTION OF THE FIGURES
[0022] FIG. 1A is a schematic plan view of a spectral-beam combiner
100 with wavelength-dispersion compensation.
[0023] FIG. 1B is an input-end view of grating 111 with
spatial-intensity graphs and wavelength-spectrum graphs of the
input laser beams.
[0024] FIG. 1C is an output-end view of grating 112 with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam.
[0025] FIG. 1D is an output-end view of output beam 90' at a
distance with a spatial-intensity graph and wavelength-spectrum
graph of the combined output laser beam.
[0026] FIG. 1E is a schematic plan view of a spectral-beam combiner
130 with wavelength-dispersion compensation.
[0027] FIG. 1F is an input-end view of grating 131 with
spatial-intensity graphs and wavelength-spectrum graphs of the
input laser beams.
[0028] FIG. 1G is an output-end view of grating 132 with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam.
[0029] FIG. 1H is an output-end view of output beam 90' at a
distance with a spatial-intensity graph and wavelength-spectrum
graph of the combined output laser beam.
[0030] FIG. 1I is a schematic plan view of a spectral beam combiner
150 with wavelength-dispersion compensation.
[0031] FIG. 1J is a schematic plan view of a spectral beam combiner
160 with wavelength-dispersion compensation using parallel
gratings.
[0032] FIG. 1K is a schematic plan view of a spectral beam combiner
161 with wavelength-dispersion compensation using non-parallel
gratings.
[0033] FIG. 1L is a schematic plan view of a spectral beam combiner
162 with wavelength-dispersion compensation using a single grating
twice and three mirrors.
[0034] FIG. 1M is a schematic plan view of a spectral beam combiner
162 with wavelength-dispersion compensation using a single grating
twice and two mirrors.
[0035] FIG. 1N is a schematic plan view of a spectral beam combiner
162 with wavelength-dispersion compensation using a single grating
twice and two mirrors.
[0036] FIG. 2A is a schematic plan view of a spectral-beam combiner
200 without wavelength-dispersion compensation.
[0037] FIG. 2B is a schematic output-end view of grating 232 with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam.
[0038] FIG. 2C is a schematic output-end view of output beam 290'
at a distance with a spatial-intensity graph and
wavelength-spectrum graph of the combined output laser beam.
[0039] FIG. 2D is a schematic representation of intensity cross
sections of input and output beams at a various distances with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam of a two-grating system of some
embodiments.
[0040] FIG. 2E is a schematic representation of intensity cross
sections of input and output beams at a various distances with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam of a one-grating system of some
embodiments.
[0041] FIG. 2F is a schematic representation of intensity cross
sections of input and output beams at a various distances for three
different input-beam shapes for a two-grating SBC system of some
embodiments.
[0042] FIG. 2G is a schematic representation of intensity cross
sections of input and output beams at a various distances for three
different input-beam shapes for a one-grating SBC system of some
embodiments.
[0043] FIG. 2H is a schematic representation of intensity cross
sections of output beams for different input-beam shapes for a
two-grating SBC system of some embodiments.
[0044] FIG. 3A is a schematic plan view of a spectral-beam-combiner
laser system 300.
[0045] FIG. 3B is a schematic plan view of an exemplary diffraction
grating system 301.
[0046] FIG. 3C is a schematic plan view of a spectral-beam-combiner
system 302 with wavelength-dispersion compensation.
[0047] FIG. 3D is a schematic plan view of a spectral-beam-combiner
laser system 304 with wavelength-dispersion compensation.
[0048] FIG. 3E is a schematic plan view of a MOPA
spectral-beam-combiner laser system 305.
[0049] FIG. 3F is a schematic plan view of a MOPA
spectral-beam-combiner laser system 306.
[0050] FIG. 3G is a schematic plan view of a MOPA
spectral-beam-combiner laser system 307 with wavelength-dispersion
compensation.
[0051] FIG. 3H is a schematic plan view of a MOPA
spectral-beam-combiner laser system 308 with wavelength-dispersion
compensation.
[0052] FIG. 3I is a schematic plan view of a spectral-beam-combiner
laser system 330 with wavelength-dispersion compensation.
[0053] FIG. 4 is a schematic block diagram of a
spectral-beam-combiner (SBC) laser system 400.
[0054] FIG. 5A is a schematic plan view of an optical
power-amplifier system 500 for use in an SBC system.
[0055] FIG. 5B is a schematic plan view of an optical master
oscillator system 550 for use in an SBC system.
[0056] FIG. 6A is a graph of an output spectrum of a seed laser
beam filtered using an angle-tuned filter.
[0057] FIG. 6B is a graph that shows high-resolution spectral
measurements of the front-end output obtained with a scanning
Fabry-Perot etalon.
[0058] FIG. 6C is a graph of measured values for one embodiment of
the LMA power amplifier's 1-micron-signal output versus power
exiting the pump-delivery fiber.
[0059] FIG. 6D is a graph of representative M.sup.2 measurements
along the horizontal-direction transverse beam axis of the 1-micron
signal, collected at an amplified average power of 100 W.
[0060] FIG. 6E is a graph of representative M.sup.2 measurements
along the vertical-direction transverse beam axis of the same
1-micron signal.
[0061] FIG. 6F is a schematic diagram of the basic test setup to
test the power-handling capabilities of a diffraction grating.
[0062] FIG. 6G shows an interferogram recorded when the heating
beam is off.
[0063] FIG. 6H shows an interferogram recorded when the heating
beam peak irradiance on the grating surface is 1.5 kW/cm.sup.2.
[0064] FIG. 6I is a schematic diagram of the layout used for a
two-channel demonstration.
[0065] FIG. 6J is a schematic diagram of the layout used for a
three-channel demonstration.
[0066] FIG. 6K is a graph of a spectrum for the 175-W two-channel
SBC-combined beam
[0067] FIG. 6L is a graph of a spectrum for the 90-W three-channel
SBC-combined beam.
[0068] FIG. 6M is a graph of M.sup.2 measurements along the
dispersed direction of the 175-W two-channel SBC-combined beam.
[0069] FIG. 6N is a graph of M.sup.2 measurements along the
non-dispersed direction of the 175-W two-channel SBC-combined
beam.
[0070] FIG. 6O is a graph of M.sup.2 measurements along the
dispersed direction for the 90-W three-channel SBC-combined
beam.
[0071] FIG. 6P is a graph of M.sup.2 measurements along the
non-dispersed direction transverse beam axis for the 90-W
three-channel SBC-combined beam.
[0072] FIG. 6Q is a graph of the dispersed-axis M.sup.2 for the
two-channel SBC-combined beam versus combined beam power.
[0073] FIG. 7A is a schematic plan view of a spectral-beam-combiner
laser system 700 with wavelength-dispersion compensation.
[0074] FIG. 7B is a flowchart of a real-time diagnostic and
adjustment process.
[0075] FIG. 7C is a schematic plan view of a spectral-beam-combiner
laser system 701 with wavelength-dispersion compensation.
[0076] FIG. 8A is a schematic plan view of a ribbon-fiber MOPA
spectral-beam-combiner laser system 800.
[0077] FIG. 8B is a schematic cross-section view of a
photonic-crystal ribbon-fiber 851.
DETAILED DESCRIPTION OF THE INVENTION
[0078] Although the following detailed description contains many
specifics for the purpose of illustration, a person of ordinary
skill in the art will appreciate that many variations and
alterations to the following details are within the scope of the
invention. Accordingly, the following preferred embodiments of the
invention are set forth without any loss of generality to, and
without imposing limitations upon the claimed invention.
[0079] In the following detailed description of the preferred
embodiments, reference is made to the accompanying drawings that
form a part hereof, and in which are shown by way of illustration
specific embodiments in which the invention may be practiced. It is
understood that other embodiments may be utilized and structural
changes may be made without departing from the scope of the present
invention.
[0080] The leading digit(s) of reference numbers appearing in the
Figures generally corresponds to the Figure number in which that
component is first introduced, such that the same reference number
is used throughout to refer to an identical component that appears
in multiple Figures. Signals and connections may be referred to by
the same reference number or label, and the actual meaning will be
clear from its use in the context of the description.
[0081] Stimulated Brillouin Scattering (SBS) can lead to power
limitations or even the destruction of a high-power fiber-laser
system due to sporadic or unstable feedback, self-lasing, pulse
compression and/or signal amplification.
[0082] One way to generate output with more controlled attributes
is to use a master-oscillator power-amplifier (MOPA) architecture.
In some embodiments, the low-power oscillator is optimized to
generate a laser seed signal having the appropriate
characteristics, such as controlled linewidth and wavelength, and
the seed signal is input to that power amplifier, which is used to
increase the output power and/or pulse energy to much higher
levels.
[0083] Recent advances in high-power fiber lasers have shown that
fiber lasers are one of the most efficient solid-state lasers that
have the capability to generate kW-order output power with very
good beam quality. The process to scale up the output power of a
single-fiber laser to a higher power level faces significant
challenges since nonlinear effects, thermal loading, fiber damage,
as well as the required pump power and brightness of pump laser
diodes (LDs) will limit the maximum output power. Several
approaches have been demonstrated to scale up output power by
combining multiple lasers. Multi-core phase-locked fiber lasers
that use the evanescent coupling between multiple cores of a fiber
to achieve coherent combining significantly reduce nonlinear
processes within the fiber core. The laser configuration is simple
and robust, but the maximum power is still limited by available
pump power and brightness of LDs as is the case in the single-fiber
system. Coherent beam combining of multiple fiber lasers using the
master-oscillator power-amplifier (MOPA) configuration solves the
power limitation, but the system is very complicated and must solve
phase-control, optical-alignment and stability issues.
[0084] Spectral beam combination (SBC) is a very promising means to
permit scaling of fiber lasers to extremely high output powers and
has achieved more than 5 watts of output power. In these
demonstrations, multiple fiber lasers that operate at slightly
different wavelengths have been multiplexed by a diffraction
grating. The beam quality and brightness can match or exceed that
achievable with coherent beam combining. Construction of reliable,
functional laser systems is much more practical with SBC than with
coherent beam combining because, while coherent beam combining
requires precise phase control to a small fraction of a wave,
spectral beam combining requires only modest bandwidth and
wavelength control of the individual sources.
[0085] In some embodiments, the present invention uses
high-efficiency gratings having multilayer dielectric layers
provided by Lawrence Livermore National Laboratory (which is
operated by the Regents of the University of California). In some
embodiments, such gratings can be made according to U.S. Pat. No.
5,907,436 entitled "Multilayer dielectric diffraction gratings"
issued May 25, 1999 to Perry et al., (incorporated herein by
reference), which is assigned to the Regents of the University of
California (Oakland, Calif.). This patent describes the design and
fabrication of dielectric grating structures with high diffraction
efficiency. The gratings have a multilayer structure of alternating
index dielectric materials, with a grating structure on top of the
multilayer, and obtain a diffraction grating of adjustable
diffraction efficiency (up to 90% or more, and, in some
embodiments, 95% or more), and variable optical bandwidth.
[0086] In some embodiments, the present invention uses
high-efficiency gratings having multilayer dielectric layers
provided by General Atomics (San Diego, Calif.). In some
embodiments, such gratings can be made according to U.S. Pat. No.
6,754,006 entitled "Hybrid metallic-dielectric grating" issued Jun.
22, 2004 to Barton et al. (incorporated herein by reference) is
assigned to General Atomics (San Diego, Calif.). This patent
describes a diffraction grating having a metallic base layer and
layers of dielectric materials of varying refractive index, where a
bottom interface of the layers is adherent to the metallic base
layer. The dielectric layers are periodically spaced on top of the
metallic base layer, leaving the metallic base layer exposed in
regions. This grating allows for the polarization insensitive
reflective properties of the base metallic layer to operate in
conjunction with the polarization sensitive diffraction properties
of the multilayer grating structure to provide near 100%
diffraction efficiency over a reasonable wavelength bandwidth,
independent of the polarization of the incident beam.
[0087] In some embodiments, the diffraction grating is a dielectric
grating chosen to provide minimal heat absorption, low
coefficient-of-thermal-expansion, and/or high
coefficient-of-thermal-conductivity, in order to reduce thermal
distortion of the diffracted beams. In some embodiments, the
dielectric grating is blazed (e.g., the angles of the sidewalls are
formed at angles parallel and perpendicular to the beam direction)
in order to increase the diffractive efficiency. In some
embodiments, the gratings have different diffraction efficiencies
for different polarizations, and the input beams are polarized in a
direction relative to the grating that maximizes the diffractive
efficiency of the grating.
[0088] FIG. 1A is a schematic plan view of a spectral-beam combiner
100 with wavelength-dispersion compensation. In some embodiments, a
plurality of input beams 96, 97, 98, . . . 99, each having a
different wavelength, are collimated into parallel beams and
impinge on a first diffraction grating 111, and each diffracts at a
different angle towards a single spot on a second diffraction
grating 112, and there each diffracts at a different complementary
angle into a single output beam 90. In some embodiments, each of
the input laser beams 96-99 has a different central wavelength
around 1060 nm, and each has an inherent spectral linewidth (e.g.,
about 0.7 nm full-width half-maximum (FWHM), in some embodiments),
so, for example, a laser beam 96 with a wavelength centered at 1060
nm would extend from 1059.65 nm to 1060.35 nm FWHM. Because the
grating 111 diffracts light at angles that are wavelength
dependent, the portion of the exemplary laser beam 96 around
1059.65 nm (the shorter-wavelength portion of the linewidth
centered at 1060 nm) would diffract at a smaller angle from the
normal vector of grating 111 ending at the upper portion of
intermediate beam 91, while the portion of the exemplary laser beam
96 around 1060.35 nm (the longer-wavelength portion of the
linewidth centered at 1060 nm) would diffract at a larger angle
from the normal vector of grating 111 ending at the lower portion
of intermediate beam 91. This is referred to herein as "chromatic
angular dispersion" or simply "chromatic dispersion." In some
embodiments, both grating 111 and grating 112 have grating lines or
grooves having substantially the same spacing, and grating 111 and
grating 112 are oriented parallel to one another in order that the
chromatic dispersion introduced by grating 111 is exactly negated
by an opposite chromatic dispersion introduced by grating 112.
Thus, output beam 90 has virtually no chromatic dispersion
introduced by gratings 111 and 112, but has much of the combined
power and intensity of the input beams in a single
diffraction-limited beam as shown by the spatial intensity graph
116. In some embodiments, the input beams are on one side of the
vector normal to the surface of grating 111, and the output beams
are on the other side of a vector that is normal to the surface of
grating 111. In other embodiments, the input and output beams are
both on the same side of the vector normal to the surface of
grating 111.
[0089] FIG. 1B is an input-end view (along section line 1B of FIG.
1A) of grating 111 shown next to spatial-intensity graphs 114 and
wavelength-spectrum graphs 115 of the plurality of input laser
beams. In some embodiments, laser beam 96 intersects grating 111 at
the top represented by the upper lump in spatial-intensity graphs
114, with successively shorter wavelength beams 97, 98, 99
intersecting grating 111 at successively lower spots. In some
embodiments, as shown in wavelength-spectrum graphs 115, the center
wavelengths and linewidths of the plurality of laser beams 96-99
are selected to slightly overlap, in order that combined spectrum
117 (see FIG. 1C), representing the sum of the input spectrums 115,
is substantially continuous over the wavelengths selected.
[0090] FIG. 1C is an output-end view (along section line 1C of FIG.
1A) of grating 112 and output beam 90 shown next to a
spatial-intensity graph 116 (representing the intensity cross
section of beam 90 in the Y direction) and wavelength-spectrum
graph 117 of the combined output laser beam.
[0091] FIG. 1D is an output-end view of output beam 90' at a
distance with a spatial-intensity graph and wavelength-spectrum
graph of the combined output laser beam. Because of the
chromatic-compensation qualities of the plurality of gratings,
output beam 90' (shown next to a spatial-intensity graph 118
(representing the intensity cross section of beam 90 in the Y
direction) and wavelength-spectrum graph 119 of the combined output
laser beam, which, in some embodiments, has an improved beam
quality M.sup.2) does not have the chromatic angular dispersion
(wavelength-dependent spreading in the Y direction) exhibited by a
spectral beam combiner using only a single grating.
[0092] FIG. 1E is a schematic plan view of a spectral-beam combiner
(SBC) 130 with wavelength-dispersion compensation. SBC 130 is
similar to spectral-beam combiner 100 of FIG. 1A, except that here,
the input and output beams are both on the same side of the vector
normal to the surface of grating 131. In some embodiments, a
plurality of input beams 96, 97, 98, . . . 99, each having a
different wavelength, are collimated into spaced-apart parallel
beams and impinge on a first diffraction grating 131, and each
diffracts at a different angle towards a single spot on a second
diffraction grating 132, and there each diffracts at a different
complementary angle into a single output beam 90. In some
embodiments, the spacings between each of the beams 96-99 and the
spacings between gratings 131 and 132 are adjusted so that all of
the beams come together at one spot on grating 132. In some
embodiments, the beam sizes are enlarged to an extent (e.g., to ten
or more millimeters diameter) that provides a sufficiently low
power density, in order to reduce heat deformation of grating 132.
The operation of system 130 is substantially similar to that of
system 100 of FIG. 1A. Thus, output beam 90 has virtually no
chromatic dispersion introduced by gratings 131 and 132, but has
much of the combined power and intensity of the input beams in a
single diffraction-limited beam as shown by the spatial intensity
graph 136. In some embodiments, the input and output beams are both
on the same side of the vector normal to the surface of grating
131. In other embodiments, the input beams are on one side of the
vector normal to the surface of grating 131, and the output beams
are on the other side of the vector normal to the surface of
grating 131, as shown in FIG. 1A.
[0093] FIG. 1F is an input-end view (along section line 1F of FIG.
1E) of grating 131 shown next to spatial-intensity graphs 134 and
wavelength-spectrum graphs 135 of the plurality of input laser
beams. In some embodiments, laser beam 96 intersects grating 131 at
the top represented by the upper lump in spatial-intensity graphs
134, with successively shorter wavelength beams 97, 98, 99
intersecting grating 131 at successively lower spots. In some
embodiments, as shown in wavelength-spectrum graphs 135, laser beam
96 has the longest wavelength, and the center wavelengths and
linewidths of the plurality of laser beams 96-99 are selected to
slightly overlap, in order that combined spectrum 137 (see FIG.
1G), representing the sum of the input spectrums 135, is
substantially continuous over the wavelengths selected.
[0094] FIG. 1G is an output-end view (along section line 1G of FIG.
1E) showing a face of grating 132 and an end-on view of output beam
90, shown next to a spatial-intensity graph 136 (representing the
intensity cross section of beam 90 in the Y direction), a
spatial-intensity graph 141 (representing the intensity cross
section of beam 90 in the X direction) and wavelength-spectrum
graph 137 of the combined output laser beam.
[0095] FIG. 1H is an output-end view of output beam 90' at a
distance with a spatial-intensity graph and wavelength-spectrum
graph of the combined output laser beam. Because of the
chromatic-compensation qualities of the plurality of gratings,
output beam 90' (shown next to a spatial-intensity graph 138
(representing the intensity cross section of beam 90' in the Y
direction), a spatial-intensity graph 142 (representing the
intensity cross section of beam 90' in the X direction) and
wavelength-spectrum graph 139 of the combined output laser beam,
which, in some embodiments, has an improved beam quality M.sup.2)
does not have the chromatic dispersion (wavelength-dependent
spreading in the Y direction) that is exhibited by a spectral beam
combiner using only a single grating (e.g., see FIG. 2A).
[0096] FIG. 1I is a schematic plan view of a spectral beam combiner
150 with wavelength-dispersion compensation using a plurality of
gratings (e.g., 151 and 152). In some embodiments, each grating is
made using conventional methods for making single gratings, for
example, such as described in U.S. Pat. No. 3,728,117 to Heidenhain
et al., U.S. Pat. No. 4,895,790 to Swanson et al., U.S. Pat. No.
6,822,796 to Takada et al., and/or U.S. Pat. No. 6,958,859 to Hoose
et al. (each of which are incorporated herein by reference). In
some embodiments, asymmetric grooves in gratings G.sub.1 151 and
G.sub.2 152 are dielectric coated, and have a groove profile and
periodicity spacing selected to maximize the efficiency of
diffracting the most power into a single-order mode (i.e., the
order that goes in the direction of the second grating) and to
minimize the power absorbed by the gratings, in order to minimize
heat distortion of the grating and to maximize output power and
intensity. In some embodiments, every input beam 96, 97, 98, . . .
99 impinges into the first grating G.sub.1 151 at the same angle
.alpha..sub.1, but each intermediate beam leaves the first grating
G.sub.1 151 at a different angle .beta..sub.11 . . . .beta..sub.1N
that depends on the wavelength of that beam, and each intermediate
beam converges to a single spot and impinges on the second grating
152 (the surface of which is parallel to the first grating 151
(G.sub.1) using the same respective angles .alpha..sub.21 . . .
.alpha..sub.1N as the outgoing angles .beta..sub.11 . . .
.beta..sub.1N for that wavelength from the first grating 151
(G.sub.1), and every beam leaves the second grating at the same
outgoing angle .beta..sub.2 in a single combined beam that is
parallel to the input beams and in the same direction.
[0097] FIG. 1J is a schematic plan view of a spectral beam combiner
160 with wavelength-dispersion compensation using parallel
gratings. In some embodiments, combiner 160 uses a first grating
G.sub.1 151 and a second grating G.sub.2 152 that are again
parallel to one another, but between which is a mirror M.sub.1 153
located at a point and angle along the converging spectral lines
such that all of the different wavelength monochromatic light beams
converge to a single location at second grating G.sub.2 152 at
angles that are the complement of the angles at which they would
impinge on second grating G.sub.2 152 if in the configuration shown
in FIG. 1I described above. That is, in some embodiments, every
input beam 96, 97, 98, . . . 99 impinges into the first grating 151
(G.sub.1) at the same angle .alpha..sub.1, but each intermediate
beam leaves the first grating 151 (G.sub.1) at a different angle
.beta..sub.11 . . . .beta..sub.1N that depends on the wavelength of
that beam, and each intermediate beam reflects from mirror M.sub.1
and converges to a single spot and impinges on the second grating
152 (the surface of which is parallel to the first grating G.sub.1
151) using the same respective angles from the normal vector from
the surface of the grating .alpha..sub.21 . . . .alpha..sub.1N as
the outgoing angles .beta..sub.11 . . . . .beta..sub.1N for that
wavelength from the first grating G.sub.1 151 (but in a
complementary angle due to the reflection by mirror M.sub.1), and
every beam leaves the second grating at the same outgoing angle
.beta..sub.2 in a single combined beam that is not parallel to the
input beams and in the complementary direction as reflected by
mirror M.sub.1 153. The combined output leaves the second grating
G.sub.2 152 same angle .beta..sub.2, but note that the output beam
is not parallel to the input beams in combiner 160.
[0098] FIG. 1K is a schematic plan view of a spectral beam combiner
161 (similar in concept to FIG. 1J) with wavelength-dispersion
compensation using non-parallel gratings. This configuration is
similar to that of FIG. 1J, but here the gratings are not parallel,
but each of the angles relative to the gratings remain the same as
the above wavelength-dispersion compensating configurations, and
mirror M.sub.1 153 is located at a point and angle along the
converging spectral lines such that all of the different wavelength
monochromatic light converges to a single location at second
grating G.sub.2 152 at angles that correspond to the angles
described above for parallel gratings. In this manner, the combined
output beam can be directed at various different output angles by
adjusting the location and angles of mirror M.sub.1 153 and second
grating G.sub.2 152, but leaving first grating G.sub.1 151 and all
the input beams in fixed locations and angles.
[0099] FIG. 1L is a schematic plan view of a spectral beam combiner
162 with wavelength-dispersion compensation using a single grating
twice and two mirrors. In this configuration, two mirrors (for
example, each at right angles to first grating G.sub.1 151 in some
embodiments), are used to redirect the converging beams to again
impinge on the first grating 151 (at location G.sub.1), wherein
each intermediate beam impinges on the grating 151 the second time
(at location G.sub.2, which in some embodiments, is separated from
location G.sub.1 on the same grating that heat from the output spot
G.sub.2, does not distort the input location G.sub.1 while in other
embodiments, location G.sub.2 and location G.sub.2 are the same
area) using the same respective angle .alpha..sub.21 as the
outgoing angle .beta..sub.21 for that wavelength from the first
grating.
[0100] FIG. 1M is a schematic plan view of a spectral beam combiner
163 with wavelength-dispersion compensation using a single grating
twice and two mirrors. Combiner 163 is similar to combiner 162
described above, but with the output beam located on the other side
of mirror M.sub.1 and closer to the input beams.
[0101] FIG. 1N is a schematic plan view of a spectral beam combiner
164 with wavelength-dispersion compensation using a single grating
twice and three mirrors. Combiner 164 is similar to combiner 162
described above, but with a third mirror such that the output beam
is located on the other side of the input beams and directed out at
an angle not parallel to the input beams.
[0102] FIG. 2A is a schematic plan view of a spectral-beam combiner
200 without wavelength-dispersion compensation. In some
embodiments, each of a plurality of input laser beams 296, 297,
298, . . . 299 are directed to a single spot on grating 232, each
coming in at a different angle that is based on the wavelength of
the input beam, in order that all the diffracted beams 291, 292,
293, . . . 294 are combined into a single output beam 290. Since
only a single grating is used, some amount of chromatic dispersion
(wavelength-dependent spreading in the Y direction) is exhibited in
the output beam 290.
[0103] In some embodiments, active real-time beam centering of each
beam is accomplished during system operation by detecting whether
the particular beam is parallel but not aligned (i.e., the beam
does not hit the single spot on the diffraction grating 232 to
which the other beams are directed) into the single output beam
290', for example, using a first detector (not shown, but similar
to detector 711 of FIG. 7A and thus closer to grating 232), e.g.,
one that receives light only if a beam is too high; and a second
detector (not shown here, but similar to detector 712 of FIG. 7A
and thus closer to grating 232), e.g., one that receives light only
if a beam is too low, both of which are connected to an
off-center-detection circuit that controls one or more mechanical
actuators to move the individual laser 256 or its optics to center
the beam on the common spot on grating 232 (see description of FIG.
7A, below). In addition, some embodiments include one or more
off-angle-detection sensors 251 and 252 connected through detector
circuit 253 to an off-angle (wavelength drift) analysis circuit 254
having outputs that control individual wavelength adjustment
circuits 255 (e.g., in some embodiments, these control, for
example, the resonant wavelength of the initial seed laser or its
output filter) for each laser 256. In some embodiments, circuit 254
has one or more inputs (not shown) that are connected to receive
pulse timing information (e.g., from pulse timing circuit for the
individual lasers), in order to determine which laser needs its
wavelength adjusted. In some embodiments, circuit 254 has outputs
(not shown) that are connected to transmit pulse timing information
(e.g., to pulse timing circuits for the individual lasers), in
order to control pulse timing and/or laser power, in order to
determine which laser needs its wavelength adjusted.
[0104] Thus, if the wavelength of one of the seed lasers drifts,
its output, while remaining centered on grating 232, will also
drift off-angle, and real-time diagnostic-and-adjustment circuit
254 detects which laser is off-angle and automatically adjusts its
wavelength until its portion of the output beam is again centered
in the far-field beam. In some embodiments, off-angle circuit 254
optionally includes an output that controls individual positioners
(e.g., in some embodiments, five-degrees-of-freedom positioners
that control, for example, X, Y, Z, pitch angle, and yaw angle) for
each laser's output (the inputs to SBC grating 232). In some
embodiments, a combination of wavelength control and positioning
control is used to keep all beams combined, parallel and aligned
into the single output beam 290' by iteratively adjusting position
of the beam on the grating (using five-degrees-of-freedom
positioners) and/or angle (using laser wavelength and/or the
five-degrees-of-freedom positioners) on each beam.
[0105] In order to reduce the effect of chromatic dispersion in
system 200, some embodiments use (for lasers 256) a plurality of
MOPA lasers having seed lasers with very narrow linewidths (e.g.,
less than one nm, in order to reduce wavelength-dependent chromatic
dispersion or spreading) and very short pulse durations (e.g.,
about ten nanoseconds or less (or, in some embodiments, 5 ns or
less), in order to prevent SBS build-up of a backward-traveling
parasitic beam in the power-amplifier fiber). Fiber lasers operated
with narrow spectral bandwidths usually suffer from stimulated
Brillouin scattering (SBS). According to the present invention,
instead of cw (continuous-wave) lasers, a fiber amplifier system is
used, seeded by a pulsed source. Some embodiments use cw diode
lasers followed by a lithium niobate Mach-Zehnder amplitude
modulator as the pulsed source (seed laser). In other embodiments,
a laser diode driven by a pulsed current source can be used if the
laser diode can be narrowband under pulsed excitation; for example,
by using a DFB (distributed feedback) laser diode. By using pulses
at least shorter than about 10 ns, SBS waves are not generated in
the gain fiber.
[0106] In some embodiments, the SBC 200 of FIG. 2A (which is
similar in some ways to FIG. 12 of U.S. Pat. No. 6,212,310 entitled
"High power fiber gain media system achieved through power scaling
via multiplexing" by Waarts et al., which is incorporated herein by
reference) is used in a system 400 as shown in FIG. 4 of the
present invention. FIG. 4 shows additional power supplies, pumps,
diagnostics and control for the lasers used to obtain input beams
296-299, in order to control wavelengths, alignments and/or timings
such that the superimposition/alignment of the beam components of
output beam 290 remain aligned (since changes in the wavelengths or
beam alignments due to heating of the laser components or heating
of grating 232 would otherwise cause the beam components to become
misaligned).
[0107] FIG. 2B is a schematic output-end view (along section line
2B of FIG. 2A) of grating 232 and output beam 290 shown next to a
spatial-intensity graph 236 (representing the intensity cross
section of beam 290 in the Y direction, which, in some embodiments,
is spread vertically by an amount based on the angles on the
incident and diffractive beams relative to the grating 232), a
spatial-intensity graph 241 (representing the intensity cross
section of beam 90 in the X direction, which, in some embodiments,
is not spread by grating 232 other than by aberrations) and
wavelength-spectrum graph 237 of the combined output laser
beam.
[0108] FIG. 2C is a schematic output-end view (along section line
2C of FIG. 2A) of output beam 290' at a distance from grating 232,
shown with two spatial-intensity graphs 241 and 236 and
wavelength-spectrum graph 237 of the combined output laser beam.
Because of the chromatic dispersion qualities of the single grating
232 (which are used to combine the beams), output beam 290' has a
certain amount of chromatic dispersion (wavelength-dependent
spreading of the beam in the Y direction) inevitably exhibited by a
spectral beam combiner using only a single grating. In some
embodiments, the amount of such chromatic dispersion is reduced by
reducing the linewidth of each input beam. e.g., to about 0.1 nm or
less. The essential purpose of the grating is to disperse or
diffract different wavelengths at different angles, such that the
different lasers, each at a different central wavelength, can be
combined by impinging at a common location on the grating and
emerging as a single beam, however real-world lasers will have a
non-zero linewidth such that the longer-wavelength end of the
linewidth from a single laser will diffract at a different angle
than the shorter-wavelength end of the linewidth (called "chromatic
dispersion" herein). In some embodiments, the individual seed
lasers are conditioned to have a very narrow linewidth (e.g., less
than 1 nm linewidth on a wavelength of 1000 to 1100 nm), in order
to minimize chromatic dispersion.
[0109] FIG. 2D is a schematic representation of intensity cross
sections of input and output beams at a various distances with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam of a two-grating system of some
embodiments. A plurality of input laser beams 271 are optically
enlarged to a 2:1 width-to-height ratio and to a size sufficient to
obtain a small enough power density on the output grating (e.g.,
less than 1500 W/cm.sub.2, in some embodiments) in order to
substantially reduce heat distortion and degradation of the output
grating. Graph 272 represents a calculated intensity profile (solid
curve) in the Y-direction (assuming a flat-top linewidth profile)
that is spread by chromatic dispersion by an amount equal to the
Y-direction FHWM width of the input beam, such that the output beam
has a 1/e.sup.2 beam height=beam width (the dotted curve shows a
Gaussian curve of a width needed such that the output beam has a
FWHM beam height=beam width). As the beam travels further from the
grating, the beam cross-sections 276 and 278 and the spectrums 277
and 279 remain substantially unchanged.
[0110] FIG. 2E is a schematic representation of intensity cross
sections of input and output beams at a various distances with a
spatial-intensity graph and wavelength-spectrum graph of the
combined output laser beam of a one-grating system of some
embodiments. At the single grating, all wavelengths of the input
beams are combined to a single Gaussian spot 281 (e.g., having a 10
mm diameter=2.omega..sub.0) having a Y-direction intensity profile
282 and an X-direction intensity profile 283. The initial beam
cross-section ("spot") is shown as 274, and its spectrum is shown
as 275. Without astigmatic optical focusing, at various distances
Z, the beam cross-section will have spread in the Y-direction by an
amount based on the linewidths of the individual input beams (e.g.,
spots 284, 286 and 288 having a spectrums 285, 287 and 289), but
with astigmatic optical focusing or compensation the spots 264, 266
and 288 will result at those distances Z. With narrower linewidths,
improved beam quality results in the various amounts of
chromatic-dispersion spreading occurring at further distances.
[0111] FIG. 2F is a schematic representation of intensity cross
sections of input and output beams at a various distances for three
different input-beam shapes for a two-grating SBC system of some
embodiments. Three sets of possible input beams are shown: a first
set of uncompensated circular beams (e.g., 10 mm by 10 mm), a
second set of 2:1 precompensated oval beams (e.g., 10 mm by 5 mm),
and a third set of 4:1 precompensated oval beams (e.g., 10 mm by
2.5 mm), with, for example, seven different center wavelengths that
are to be SBC combined. At a first grating spacing, for example,
the uncompensated circular beams would form an elongated spot
(e.g., 10 mm by 18.7 mm), while the precompensated beams would have
heights equal to widths (selected as the FWHM dimensions or the
1/e.sup.2 dimensions, as desired). As the distance from the second
output grating increases, these beam intensity cross sections
substantially do not change due to chromatic dispersion since the
two-grating configuration stops further chromatic dispersion after
the second grating.
[0112] FIG. 2G is a schematic representation of intensity cross
sections of input and output beams at a various distances for three
different input-beam shapes for a one-grating SBC system of some
embodiments. Three sets of possible input beams are shown: a first
set of circular beams (e.g., 1 mm by 1 mm), a second set of
circular beams (e.g., 10 mm by 10 mm), and a third set of circular
beams (e.g., 30 mm by 30 mm), with, for example, seven different
center wavelengths that have been SBC combined at one spot on the
single grating (the output grating). At a first distance from the
grating, for example, the small circular beams would form an
elongated spot with a very bad aspect ratio (e.g., 1 mm by 9.73
mm), while the larger initial beams would have heights closer to
widths (e.g., 10 mm by 18.73 mm or 30 mm by 38.73 mm, respectively,
selected as the FWHM dimensions or the 1/e.sup.2 dimensions, as
desired). As the distance from the second output grating increases,
these beam intensity cross sections do substantially change due to
chromatic dispersion since the one-grating configuration does not
further chromatic dispersion after the single grating. In some
embodiments, an astigmatic focusing element (such as a cylindrical
lens or mirror of an appropriately selected shape, see description
of FIG. 3F below) is used to reshape the output beam so its
height-to-width ratio stays at about one.
[0113] FIG. 2H is a schematic representation of intensity cross
sections of output beams for different input-beam shapes for a
two-grating SBC system of some embodiments. The cross-section
intensity profile of a circularly symmetric Gaussian beam is shown
at the left, next to a plan view of the same cross-section, showing
the full-width half-maximum (FWHM) and 1/e.sup.2 circumferences.
Next are shown the same graphs for a beam with 1.5:1
precompensation, a beam with 2:1 precompensation, a beam with 3:1
precompensation, and a beam with 4:1 precompensation, each set so
its 1/e.sup.2 width and its 1/e.sup.2 height are approximately
equal. Beneath the 1/e.sup.2 graphs for the 3:1 precompensation are
graphs for 3:1 precompensation where the FWHM width is set equal to
the FWHM height.
[0114] FIG. 3A is a schematic plan view of an exemplary
spectral-beam-combiner laser system 300. In various embodiments,
both power oscillator (PO) and master oscillator power amplifier
(MOPA) configurations are used to combine the wavelengths from
different emitters into a single beam. FIG. 3A shows an exemplary
power-oscillator spectral-beam-combined fiber laser system 300
where a partially reflecting output mirror 329 (the output coupler)
provides feedback through grating 310 (G.sub.0) to the plurality of
different fiber lasers 325 through 326 and locks each one to a
wavelength that allows for a single output beam 380. In some
embodiments, a spatial filter (not shown) is included to prevent
crosstalk between channels. Each fiber laser 325-326 receives
feedback at a slightly different center wavelength from the output
coupler 329 and grating 310 that forces its operation at a unique
center wavelength such that its output beam 386-389 spatially
overlaps that of the other fiber lasers at grating 310 (G.sub.0)
and through output coupler 329. The output beam 380 has chromatic
dispersion due to the non-zero linewidth of each laser beam and the
spread of diffraction angles that linewidth incurs at grating 310
(G.sub.0), which acts both to determine each fiber laser's
operating wavelengths, and to combine their output beams. FIG. 3E
shows fiber MOPAs combined using a similar optical system 305, but
with the wavelengths of each respective master oscillator 351 . . .
356 set to overlap the beam from each other fiber amplifier in the
far-field of beam 290. In some embodiments, the master oscillators
use tunable sources set to the wavelength required for each
channel, or an array of fibers or diodes in a power oscillator SBC
system similar to that shown in FIG. 3A.
[0115] In some embodiments, system 300 includes a plurality of
lasers (325, 326, and the like, each operating at a slightly
different peak wavelength) that are spectrally combined using
grating G.sub.0 310 and output through partial reflector 329. Each
fiber laser (325, 326, and the like) of the plurality includes a
laser cavity having the fiber laser gain medium (336, 339 and the
like) having backside M.sub.b reflector (331, 335, and the like)
and signal input/output port (334, 337, and the like), transform
lens 321, grating 310, and output coupler (partially transmissive
mirror 329), wherein the diffraction angle of the grating 310
determines the operating wavelengths of each fiber laser in the
fiber-laser array. For example, the first laser 325 operates at
wavelength .lamda..sub.1 and the N.sup.th laser 326 operates at
wavelength .lamda..sub.N. In some embodiments, each fiber laser
(325, 326, and the like) receives feedback from the output coupler
(partially transmissive mirror 329) that forces its operation at a
wavelength such that its output beam spatially overlaps the beams
of the other fiber lasers (325, 326, and the like) and has a
wavelength that is spectrally next to the wavelengths of the lasers
to either side. Each fiber laser (325, 326, and the like) has a
back-end reflector (e.g., a reflective grating or mirror M.sub.b or
other suitable feedback device, designated 331 and 335). Each fiber
laser (325, 326, and the like) has a suitable pump mechanism, such
as a double-clad fiber pumped by a high-power laser diode stack.
The grating 310 acts both to determine the fiber laser operating
wavelengths and to combine their output beams into beam 380. In
this description, a model that analyzes dependence of laser
linewidth on beam quality of an SBC system is given. Based on the
model, a novel configuration (having a plurality of gratings) that
significantly improves beam quality is discussed. The experimental
results have shown that control of fiber-laser linewidth (i.e.,
making the linewidth very narrow) is one key to achieving
high-power single-grating SBC laser systems.
[0116] In other embodiments, other types of lasers (such as
optically pumped semiconductor lasers (OPSLs), optically pumped
photonic-crystal fiber lasers, and the like) are substituted for
fiber lasers 325, 326, and the like in system 300 (i.e., while the
rest of the apparatus remains substantially the same, the
fiber-gain medium 336, 339, and the like are replaced with
semiconductor gain media, photonic-crystal fiber gain media, and
the like).
[0117] In some embodiments, the number of grooves on the grating
illuminated by a beam determines the spectral resolving power of
the grating: R = .lamda. .DELTA. .times. .times. .lamda. = mN
##EQU1## where m is the diffraction order and N is the number of
grooves illuminated. For most cases of interest, we use gratings
with a single diffraction order, m=-1, and N can be determined from
the beam diameter, D and the angle of incidence, .alpha.. D = 2
.times. .omega. cos .times. .times. .alpha. = Nd = N g ##EQU2##
where d is the spacing between grooves (in units of length) or g is
the groove density. Then the wavelength spread that can be resolved
by the grating is .lamda. .DELTA. .times. .times. .lamda. = N = 2
.times. .omega. d .times. .times. cos .times. .times. .alpha.
.times. .times. or .times. .times. .DELTA..lamda. = .lamda. .times.
.times. d .times. .times. cos .times. .times. .alpha. 2 .times.
.omega. ##EQU3## For various grating groove densities, examples can
be calculated of the beam sizes required to stay below a given
irradiance (i.e., to avoid damage) and the corresponding wavelength
spread. spot .. .times. width = .omega. = cos .times. .times.
.alpha. .pi. .times. POWER POWER .times. / .times. AREA .times.
.times. and ##EQU4## .times. grating .. .times. resolution =
.lamda. n + 1 - .lamda. n = .lamda. .times. .times. cos .times.
.times. .alpha. g .times. 2 .times. .omega. = .lamda. .times.
.times. cos .times. .times. .alpha. g .times. 2 .times. cos .times.
.times. .alpha. .pi. .times. POWER POWER .times. / .times. AREA
##EQU4.2## For many gratings, high efficiency is obtained at angles
around the Littrow angle. Therefore for the example below, we
assume a single incidence angle equal to the Littrow reflection
angle. As example irradiance values, we use 1.5 kW/cm.sup.2 and 15
kW/cm.sup.2 for the beam areas on the grating (including the factor
for the Littrow angle of incidence), and output powers of 1 and 100
kW. Since the higher irradiance value of 15 kW/cm.sup.2 is still
well below typical damage thresholds of dielectric coatings, we
believe this value should be attainable with appropriate selection
of the materials used in the dielectric coatings and substrate,
along with consideration of grating mounting and cooling
configurations for very high power beams.
[0118] One of the advantages of high-power fiber lasers over
traditional solid-state lasers is that diffraction-limited beam
quality can be generated at kW output power levels as the beam
quality is determined by the core numeric aperture (NA) and the
mode-field diameter and thus is extremely insensitive to the output
power. When the output beams of a plurality of fiber lasers with
single-mode cores are combined by the SBC, it is desirable for the
combined beam to have the same beam quality as the individual
lasers if all lasers operate at the resonator-determined linewidth
and there is no additional aberrations induced by the optical
components. However the linewidth of each laser can be broadened by
non-linear processes, such as stimulated Brillouin scattering (SBS)
and self-phase-modulation, as well as gain saturation occurring in
inhomogeneous gain media. As a result, the majority of high power
fiber lasers operating at greater than 100 Watts have linewidths
greater than one nm. To understand how the laser linewidth affects
the beam quality of the SBC, we have developed a simple model for
some embodiments in which the linewidth broadening is taken into
account.
[0119] FIG. 3B is a schematic plan view of an exemplary diffraction
grating system 301 that includes a grating 310. Consider a
single-mode Gaussian beam with a linewidth .DELTA..lamda. incident
on diffraction grating 310 as shown in FIG. 3B. Assuming the
Rayleigh length of the beam is long enough in the range of interest
(in some embodiments, that is true when the beam diameter is more
than 10 mm), then the beam quality of the diffracted beam can be
described by M 1 2 = .omega. 1 .times. .theta. 1 .omega. 0 .times.
.theta. 0 EQ . .times. ( 1 ) ##EQU5## where, .omega..sub.0 and
.theta..sub.0 are the beam waist (radius) and divergence of the
incident beam (assumed, in some embodiments, to be a perfect
Gaussian) respectively, while .omega..sub.0 and .theta..sub.1 are
the beam waist and divergence of the diffracted beam
respectively.
[0120] The ratio of the width of a collimated diffracted beam to
that of a collimated incident beam is .omega. 1 .omega. 0 = cos
.times. .times. .beta. 1 cos .times. .times. .alpha. 1 EQ . .times.
( 2 ) ##EQU6## Here .alpha..sub.1 and .beta..sub.1 are the angles
of incidence and diffraction, respectively.
[0121] Accordingly, the ratio of the divergence of a collimated
diffracted beam to that of a collimated incident beam is .theta. 1
.theta. 0 = cos .times. .times. .alpha. 1 cos .times. .times.
.beta. 1 EQ . .times. ( 3 ) ##EQU7## However, equation (3) does not
account for angular spread due to the linewidth .DELTA..lamda. of
the beam. To simplify the analysis, assuming a top-flat spectral
profile within the linewidth, the full angular width of the
diffracted beam induced by linewidth broadening is .theta. B = g
.times. .times. .DELTA..lamda. cos .times. .times. .beta. 1 EQ .
.times. ( 4 ) ##EQU8## where g is the grating groove density. When
this angular spread is included, the diffracted beam divergence
becomes .theta. 1 = cos .times. .times. .alpha. 1 cos .times.
.times. .beta. 1 .times. .theta. 0 + 1 2 .times. .theta. B EQ .
.times. ( 5 ) ##EQU9## and then the beam quality of the diffracted
beam can be described by M 1 2 = .theta. 1 .times. cos .times.
.times. .beta. 1 .theta. 0 .times. cos .times. .times. .alpha. 1 =
1 + .theta. B .times. cos .times. .times. .beta. 1 .theta. 0
.times. cos .times. .times. .alpha. 1 = 1 + g .times. .times.
.DELTA..lamda. .theta. 0 .times. cos .times. .times. .alpha. 1 EQ .
.times. ( 6 ) ##EQU10## For the SBC cavity shown in FIG. 3A, if
there is no additional linewidth broadening within the system, and
the laser operates at a linewidth defined by the cavity resolution:
.DELTA. .times. .times. .lamda. = d gf .times. cos .times. .times.
.alpha. 1 EQ . .times. ( 7 ) ##EQU11## where f is the focal length
of the collimation lens and d is the fiber mode field diameter.
Then Equation (6) becomes M 1 2 = .theta. 1 .times. cos .times.
.times. .beta. 1 .theta. 0 .times. cos .times. .times. .alpha. 1 =
1 + d 2 .times. .times. f .times. .times. .theta. 0 .apprxeq. 2 EQ
. .times. ( 8 ) ##EQU12##
[0122] Equation (8) shows that the beam quality of the SBC laser
will become worse than each original laser if the laser linewidth
of each laser is equal to the cavity resolution defined by Equation
(7). For conventional solid-state laser systems, the laser
linewidth is always narrower than the cavity defined linewidth as
the gain saturation clamps the lasing linewidth in a homogenous
broadening gain media. However, high-power fiber lasers operate in
a different regime as both homogenous and inhomogenous broadening
contribute to linewidth broadening. Nonlinear processes broaden the
linewidth. This is why the majority of high-power fiber lasers
produce a broadband output.
[0123] FIG. 3C is a schematic plan view of a spectral-beam-combiner
system 302 with wavelength-dispersion compensation. To reduce the
impact of the linewidth broadening of the single grating of FIG. 3B
on the beam quality, some embodiments use a two-grating
configuration 302, for example, as shown in FIG. 3C. In some
embodiments, the second grating 311 is identical to the first
grating 310, and is used to compensate the angular spread of the
beam due to linewidth broadening (also called herein "chromatic
dispersion") by introducing an equal and opposite angular spread.
In terms of the grating equation, it is very easy to understand
that the diffracted beam from the second grating must be parallel
to the incident beam, i.e.,
.theta..sub.2=.theta..sub.0;.alpha..sub.2=.beta..sub.1;.beta..sub.2=.alph-
a..sub.1; EQS.(9) The beam size of the diffracted beam after the
second grating is thus .omega. 2 = cos .times. .times. .beta. 2 cos
.times. .times. .alpha. 2 .times. ( .omega. 1 + .theta. 1 .times. l
) EQ . .times. ( 10 ) ##EQU13## where l is the length of optical
path between the two gratings. Then the beam quality of the
diffracted beam after the second grating is ( M 2 ) 2 = .omega. 2
.times. .theta. 2 .omega. 0 .times. .theta. 0 = 1 + .theta. 0
.times. l .omega. 0 .times. ( M 1 ) 2 EQ . .times. ( 11 ) ##EQU14##
If two gratings are placed very close, then it is possible to
simplify equation (11) into (M.sub.2).sup.2.apprxeq.1 EQ.(12) by
using l=.omega..sub.0, and .theta..sub.0<<1.
[0124] FIG. 3D is a schematic plan view of a spectral-beam-combiner
laser system 304 withwavelength-dispersion compensation based on
the two-grating scheme, as used in some embodiments to provide a
modified and improved SBC fiber laser system. In some embodiments,
system 304 includes three gratings (G.sub.0 310, G.sub.1 311, and
G.sub.2 312), a fiber-laser array, cavity mirrors M.sub.b and
M.sub.f, and one or more transform lenses 321, which together form
power oscillator system 319. In some embodiments, system 304
includes a plurality of lasers (327, 328, and the like, each
operating at a slightly different peak wavelength) that are
spectrally combined using grating G.sub.0 310 (a single grating
that is used to set each of the different wavelengths), reflected
by highly reflective reflector 322 back to grating 310 that again
separated the various wavelengths to their respective laser fibers.
In some embodiments, system 304 includes a plurality of
active-gain-media optical fibers 336 . . . 339 each capable of
lasing across a range of wavelengths, but each forced to lase at
one of a plurality of slightly different wavelength .lamda..sub.1
386 . . . .lamda..sub.N 389, respectively, as set by different
diffraction angles to grating G.sub.0 310. In some embodiments, the
ends 334 . . . 337 of the fibers are laterally spaced such that
collimating lens 321 directs each laser beam toward grating 310 at
a different successive angle.
[0125] Each fiber laser (327, 328, and the like) of the plurality
includes a laser cavity having the fiber laser gain medium (336,
339 and the like) having a partially reflecting reflector and
output coupler (e.g., front-side mirror) M.sub.f (332, 333, and the
like) and signal input/output port (334, 337, and the like),
transform lens 321, grating 310, and highly reflective mirror 322,
wherein the diffraction angle of the grating 310 determines the
operating wavelengths of each fiber laser in the fiber-laser array.
For example, the first laser 327 operates at wavelength
.lamda..sub.1 and the N.sup.th laser 328 operates at wavelength
.lamda..sub.N. In some embodiments, each fiber laser (327, 328, and
the like) receives wavelength-selective feedback from the backside
mirror 322 and grating 310 that forces its operation at a
wavelength such that its output beam, after diffraction by gratings
311 and 312, spatially overlaps the beams of the other fiber lasers
(327, 328, and the like) to form combined output beam 390, and has
a wavelength that is spectrally next to the wavelengths of the
lasers to either side. The back mirror M.sub.b 322 has a very high
reflection at all wavelengths of the laser array. The left-side
input facets of the fiber array and the grating G.sub.0 are located
at the back and front focal planes of the lens F.sub.0,
respectively. With this arrangement, the laser beams from the laser
array are collimated and overlapped on the grating 310 (G.sub.0).
The operation wavelength of each laser is determined by the optical
dispersion provided by the grating 310, the lens 321 (F.sub.0) and
the top-to-bottom spacing of fiber input facets 334 and 337. On the
output side, there are two identical gratings G.sub.1 and G.sub.2
as well as a collimating lens array.
[0126] The multiple laser beams from the laser array are collimated
by the lens array f.sub.1-f.sub.N 341 into substantially parallel
beams that illuminate the face of the first grating G.sub.1 at the
same incident angle. After the diffraction by the grating G.sub.1,
all the beams have slightly different diffraction angles, since the
wavelengths from the laser array are slightly different. As a
result, all laser beams can be completely overlapped at a certain
position where the grating G.sub.2 is located. These overlapped
beams are then diffracted by the grating G.sub.2 at the same
diffraction angle (this can be adjusted by changing the
top-to-bottom spacing of each individual right-side end 332-333 of
the fiber array, and by changing the distance between gratings 311
and 312). Because the two gratings 311 and 312 (G.sub.1 and
G.sub.2) have the same grating frequency (e.g., line spacings of
between about 600 lines per mm to 1800 lines per mm, in some
embodiments) and are parallel to each other, the overlapped output
beams 390 diffracted by G.sub.2 must be parallel to the incident
beams 396-399 impinging on the grating G.sub.1. The
partial-reflection mirror M.sub.f is the output coupler for each
fiber laser, and, in some embodiments, is coated on the fiber facet
and provides feedback and transmission to each respective one of
the laser beams. By this way, many lasers operating at different
wavelengths can be combined into one beam without beam-quality
reduction.
[0127] Further, unlike system 300 of FIG. 3A in which the chromatic
dispersion angle of the output beam increases if the linewidths of
the individual lasers are broadened (or decreases if the linewidths
of the individual lasers are narrowed), the individual lasers
327-328 of FIG. 3D can each have quite broad linewidths and the
compensating gratings prevent any increase in chromatic dispersion.
Therefore, the device can increase the total laser power by a
factor of N if there are N laser gain media in the system,
regardless of the linewidths of the input laser beams.
[0128] In other embodiments, other types of lasers (such as
optically pumped semiconductor lasers (OPSLs), optically pumped
photonic-crystal fiber or rod lasers, and the like) are substituted
for fiber lasers 327, 328, and the like in system 304 (i.e., while
the rest of the apparatus remains substantially the same, the
fiber-gain medium 336, 339, and the like are replaced with
semiconductor gain media, photonic-crystal fiber or rod gain media,
and the like).
[0129] With some embodiments of the two-grating approach 304 as
shown in FIG. 3D, the beam quality of the combined beam is
substantially insensitive to linewidth of the individual input
lasers. Therefore, in some embodiments, each laser of system 304
can operate either with narrow linewidths (e.g., using very short
pulses to avoid SBS buildup) or at relatively wide linewidths at
high power, which also helps avoid the detrimental SBS process that
otherwise amplifies a backward-traveling wave, draining pump power
and reducing output power and efficiency. In contrast, note that
system 300 of FIG. 3A requires, or benefits greatly if the
linewidth of each input laser is very narrow, in order that the
chromatic dispersion angle for each component of the output beam is
minimized.
[0130] FIG. 3E is a schematic plan view of a MOPA
spectral-beam-combiner laser system 305. In some embodiments,
system 305 includes a plurality of master oscillators (MOs) 351 . .
. 356. MO 351 includes a partial-reflecting diffraction Bragg
reflector 352 in one end of a laser fiber, and a facet 354 at the
other (left) end, which transmits an expanding beam 386 that is
somewhat collimated by lens 321 onto diffraction grating 310, which
diffracts the beam at substantially the Littrow angle to mirror
322, where it is reflected back along the identical path in
reverse, being focussed and received into facet 354. This laser
oscillates at a characteristic wavelength set by the angles at
which the beam intersects grating 310 (e.g., a somewhat longer
wavelength than that of laser 356, which includes grating 357 and
facet 359, but otherwise functions in a similar manner. Master
oscillator 351 provides as its output a seed laser signal of a very
narrow linewidth through one-way optical isolator 353 into power
amplifier 355, which, in some embodiments, is a large-mode-area
(LMA) amplifying fiber, PC fiber, PC rod, or the like having a low
NA and operating substantially in the fundamental mode (for
example, by methods described in U.S. Pat. No. 5,818,630 to
Fermann, et al., entitled "Single-mode amplifiers and compressors
based on multi-mode fibers", or U.S. Pat. No. 6,496,301, to Koplow,
et al., entitled "Helical fiber amplifier", which are incorporated
herein by reference, such that only the single mode exits through
the end facet 362 of the endcap. In some embodiments, the end
facets 354, 359, 362, and 363 are cleaved or polished at an angle
chosen to minimize feedback and maximize the emitted light. In a
similar manner, master oscillator 356 provides as its output a seed
laser signal of a very narrow linewidth through one-way optical
isolator 358 into power amplifier 360, which emits its output
through facet 363. In some embodiments, a large plurality of other
similar MOPAs, each operating at a different intermediate
wavelength are implemented between MOPA 351, 355 and MOPA 356, 360,
and all the output beams are somewhat collimated and directed to
grating G.sub.1 311 at corresponding angles to those of grating
G.sub.0 310 and spectrally combined into a single output beam 290.
In some embodiments, a larger v while a smaller spot is used on
grating G.sub.1 311 in order that fewer grating lines are used and
a wider bandwidth is obtained.
[0131] FIG. 3F is a schematic plan view of a MOPA
spectral-beam-combiner laser system 306. System 306 is
substantially identical to system 305 described above, but with the
addition of a concave astigmatic mirror 369 (one having a negative
cylindrical shape, wherein the curved cross section is a parabola
or other shape useful for re-collimating the chromatic dispersion)
used with convex astigmatic mirror 368 (one having a positive
cylindrical shape, wherein the curved cross section is a parabola
or other shape useful for re-collimating the chromatic dispersion)
together used to re-collimate the linewidth-dispersion angle and
maintain a better (lower) M.sup.2 value.
[0132] FIG. 3G is a schematic plan view of a MOPA
spectral-beam-combiner laser system 307 with wavelength-dispersion
compensation. In some embodiments, a plurality of laser sources
(such as MOPA.sub.1-MOPA.sub.N) are provided in a manner similar to
FIG. 3F, except that lens 361 F.sub.1 forms substantially parallel
intermediate collimated beams directed to grating(s) 311, and, in
some embodiments, a plurality of separate gratings 311 are
provided, each parallel to output grating 312, but each of the
plurality of separate gratings 311 located at a distance from
output grating 312 such that the beam(s) it directs are
substantially the same size on output grating 312 after linewidth
spreading. (Otherwise, the beams that intersect grating 312
centered on the same spot, but with different amounts of spectral
spreading due to the different distances for the beams and
different angles on intersection at grating 312.)
[0133] FIG. 3H is a schematic plan view of a MOPA
spectral-beam-combiner laser system 308 with wavelength-dispersion
compensation. System 308 is similar to system 307. In some
embodiments, a plurality of laser sources (such as
MOPA.sub.1-MOPA.sub.N) are provided in a manner similar to FIG. 3G,
except that a plurality of separate lenses 368 F.sub.1-F.sub.N form
substantially parallel intermediate collimated beams directed to
grating(s) 311.
[0134] Certain experiments have been performed. In some
embodiments, lasers used in these experiments are commercial fiber
lasers with a maximum output power of 30 W. In some embodiments,
the Yb-doped double-clad fibers are pumped by fiber-coupled LDs via
star-couplers. In some embodiments, the fibers are
polarization-maintaining single mode fibers with a
mode-field-diameter of 6.5 .mu.m. In some embodiments, each fiber
laser has a high-reflection fiber-Bragg reflector (FBG) with
reflection >99% attached as the rear reflector and has a
bandwidth of 7 nm at a central wavelength of 1078 nm. On the output
side, the fibers are angle-polished and mounted on nano-positioning
stages to construct the SBC cavity.
[0135] The results obtained in this study show that in some
embodiments, SBC is a promising method for beam combination of
high-power fiber lasers. Based on the model, in some embodiments,
control of laser linewidth and avoiding grating heating distortion
are the keys to achieving diffraction-limited beam quality. The
laser linewidths of the fiber lasers in a single-grating
configuration must be somewhat narrower than the resolution of the
SBC cavity. Because commercial high power fiber lasers have broad
linewidth, a new SBC configuration with two or three gratings is
needed to combine them to achieve very high power and preserve beam
quality. Furthermore, a high-efficiency diffraction grating with
high damage threshold and ultra-low thermal distortion will be
needed in any configuration. If these issues are properly
addressed, power scaling to the multi-kilowatt regime using many
fiber lasers will be feasible.
[0136] FIG. 31 is a schematic plan view of a spectral-beam-combiner
laser system 330 with wavelength-dispersion compensation. System
330 uses substantially the same output SBC as system 304 of FIG. 3D
(see description for FIG. 3D above for those aspects), but uses a
different method and apparatus for setting successively different
lasing wavelengths of the lasers. In some embodiments, system 330
includes three gratings (G.sub.0' 309, G.sub.1 311, and G.sub.2
312), a fiber-laser array, and cavity mirrors M.sub.f, which
together form power oscillator system 318. In some embodiments,
system 330 includes a plurality of lasers (323, 324, and the like,
each operating at a slightly different peak wavelength) that each
cross grating G.sub.0' 309 at a slightly different angle (grating
G.sub.0' 309 is a single grating or single grating-forming process
that is used to set each of the different wavelengths), wherein
grating G.sub.0' 309 forms a highly reflective Bragg reflection
grating on each fiber to reinforce the various wavelengths to their
respective laser fibers. In some embodiments, system 330 includes a
plurality of active-gain-media optical fibers 336 . . . 339 each
capable of lasing across a range of wavelengths, but each forced to
lase at one of a plurality of slightly different wavelength
.lamda..sub.1 386 . . . .lamda..sub.N 389, respectively, as set by
different angles of each fiber to grating G.sub.0' 309.
[0137] Each fiber laser (323, 324, and the like) of the plurality
includes a laser cavity having the fiber laser gain medium (336,
339 and the like) having a partially reflecting reflector and
output coupler (front-side mirror) M.sub.f (332, 333, and the
like). Each uses a wavelength-determining highly reflective Bragg
reflection grating G.sub.0' 309, wherein the different angles of
each fiber to grating G.sub.0' 309 determine the operating
wavelengths of each fiber laser in the fiber-laser array. For
example, the first laser 323 operates at wavelength .lamda..sub.1
and the N.sup.th laser 324 operates at wavelength .lamda..sub.N. In
some embodiments, each fiber laser (323, 324, and the like)
receives wavelength-selective feedback from the backside Bragg
reflection grating G.sub.0' 309 that forces its operation at a
wavelength such that its output beam, after diffraction by gratings
311 and 312, spatially overlaps the beams of the other fiber lasers
(323, 324, and the like) to form combined output beam 390, and has
a wavelength that is spectrally next to the wavelengths of the
lasers to either side. On the output side, there are two identical
gratings G.sub.1 311 and G.sub.2 312 as well as a collimating lens
array f.sub.1-f.sub.N 341.
[0138] The multiple laser beams from the laser array are collimated
by the lens array f.sub.1-f.sub.N 341 into substantially parallel
beams that illuminate the face of the first grating G.sub.1 at the
same incident angle, as described above for FIG. 3D. Further, in
some embodiments, unlike system 300 of FIG. 3A in which the
chromatic dispersion angle of the output beam increases if the
linewidths of the individual lasers are broadened (or decreases if
the linewidths of the individual lasers are narrowed), the
individual lasers 327-328 of FIG. 3D can each have quite broad
linewidths and the compensating gratings prevent any increase in
chromatic dispersion. Therefore, the device can increase the total
laser power by a factor of N if there are N laser gain media in the
system.
[0139] In other embodiments, other types of lasers (such as
optically pumped semiconductor lasers (OPSLs), optically pumped
photonic-crystal fiber, rod, or rod-like fiber lasers, and the
like) are substituted for fiber lasers (323, 324, and the like) in
system 330 (i.e., while the rest of the apparatus remains
substantially the same, the fiber-gain medium 336, 339, and the
like are replaced with semiconductor gain media, photonic-crystal
fiber gain media, and the like).
[0140] FIG. 4 is a schematic block diagram of a
spectral-beam-combiner (SBC) laser system 400. In some embodiments,
system 400 includes a plurality of seed lasers (master oscillators)
whose output wavelength, line width, power, pulse width, pulse
timing, and the like can be individually and/or collectively
controlled by power supplies, diagnostic, and control module 450.
The outputs of the master oscillators 410 are amplified by optical
power amplifiers 420, such that blocks 410 and 420 together form an
array of MOPAs. The outputs of the power amplifiers 420 are
combined into one or more single beams by one or more spectral beam
combiners 430. The one or more single beams are then optionally
pointed by one or more output-beam directional pointers 440 into
one or more output beams 441. In some embodiments, one or more
signals or beam information data 461 are obtained from the output
beam and returned as inputs to block 460. In some embodiments, the
output beam(s) 441 are very high power (e.g., where the SBC 430 is
used to achieve higher powers than otherwise available) and used
for military purposes such as antiaircraft, antitank, antiship,
antisatellite, antipersonnel, antistructure, and the like. In some
such embodiments, block 460 represents a circuit that does
automatic threat analysis, target identification, and
countermeasures control, and block 440 provides beam direction
control.
[0141] In other embodiments, the output beam(s) 441 are very high
power and used for industrial purposed such as welding or cutting
of metal, glass, ceramic or plastic, cutting of wood or fabric,
annealing of materials such as metal, semiconductors and the like,
and other industrial purposes. In some such embodiments, block 460
represents a circuit that does pattern and material analysis
(determining what has been cut or welded and what needs to be cut
or welded), and block 440 provides beam direction and/or
work-piece-movement control.
[0142] In yet other embodiments, the output beam(s) 441 are
relatively low power and used for machine vision or medical
illumination and/or scanning purposes (e.g., where the SBC 430 is
used to achieve broader optical spectra with laser-source-type
intensities than otherwise available), such as optical coherence
tomography (for example, as described in U.S. Pat. No. 6,950,692
entitled "Optical coherence tomography apparatus, optical fiber
lateral scanner and a method for studying biological tissues in
vivo", U.S. Pat. No. 6,950,692 entitled "Optical coherence
tomography apparatus, optical fiber lateral scanner and a method
for studying biological tissues in vivo" and U.S. Pat. No.
6,882,431 entitled "Quantum optical coherence tomography data
collection apparatus and method for processing therefor," each of
which is incorporated herein by reference).
[0143] In still other embodiments, the output beam(s) 441 are
relatively low power and used for telecommunications purposes such
as wavelength-division multiplexing (WDM) or dense
wavelength-division multiplexing (DWDM) (e.g., where the SBC 430 is
used to either combine a large plurality of narrow linewidth
different wavelength modulated signals, and the system also
optionally includes an SBC with wavelength-dispersion compensation
that is used to separate a single incoming optical communications
signal beam into a plurality of separate beams of the different
component signals).
[0144] FIG. 5A is a schematic plan view of an optical
power-amplifier system 500 for use in a MOPA-SBC system. In some
embodiments, system 500 includes a plurality of seed lasers 550
(e.g., each outputting 5 ns pulses at a repetition rate of 10 MHz
(i.e., a repeating series of 5 ns on, 95 ns off)) as the master
oscillators, where the output beams are collimated by respective
lenses 511 (one for each master oscillator, in some embodiments),
are isolated by one or more one-way optical isolators 514 and
optionally 515, and enter the corresponding power amplifier 532
through dichroic beamsplitter 520 and focusing lens 530. In some
embodiments, power amplifier fiber 532 is a large-mode-area,
dual-clad fiber amplifier that is pumped by one or more arrays of
pump laser diodes (LDs, not shown) whose light (in some
embodiments, about 400 watts or more) enters through fiber 528 to
fiber end 526, and through objective lens 524 and off highly
reflective mirror 522 towards and reflected by dichroic
beamsplitter 520 into the power amplifier fiber 532. The output
signal beam and residual pump light exit the fiber 523 towards the
right, and are collimated by lens 534 towards dichroic beamsplitter
536, where the residual waste pump light 538 is reflected downward,
and the signal 540 is transmitted. In some embodiments, a
power-amplifier system 500 is used for each of the power amplifiers
420 of FIG. 4.
[0145] In some embodiments, realizing diffraction-limited Spectral
Beam Combining (SBC) of multiple high-power fiber lasers requires
that each laser possess a spectral width on the order of a few GHz
(i.e., a very narrow linewidth; see, e.g., A. Liu, R. Mead, T.
Vatter, A. Henderson and R. Stafford, "Spectral beam combining of
high power fiber lasers," in Proceedings of SPIE, vol. 5335, 81-88
(2004), incorporated herein by reference) and exhibit
diffraction-limited spatial output. Obtaining high-efficiency beam
combining and ensuring amplitude stability for the combined beam
additionally requires, due to the polarization sensitivity typical
of high-power diffraction gratings (see, e.g., M. D. Perry, R. D.
Boyd, J. A. Britten, D. Decker, B. W. Shore, C. Shannon, E. Shults,
and L. Li, "High-efficiency multilayer dielectric diffraction
gratings," Opt. Lett., vol. 20, 940-942 (1995) (and U.S. Pat. No.
5,907,436 entitled "Multilayer dielectric diffraction gratings"
issued May 25, 1999 to Perry et al., which are incorporated herein
by reference); and Karl Hehl, J. Bischoff, U. Mohaupt, M. Palme, B.
Schnabel, L. Wenke, R. Bodefeld, W. Theobald, E. Welsch, R.
Sauerbrey and H. Heyer, "High-efficiency dielectric reflection
gratings: design, fabrication, and analysis," Appl. Opt., vol. 38,
6257-6271 (1999); incorporated herein by reference) that each laser
in the SBC array have a polarization extinction ratio (PER)
>>1 and a stable polarization orientation with respect to
time. One architecture for MOPAs developed at Aculight (assignee of
the present invention) to meet these needs is illustrated in FIG.
5A. In some embodiments, the power amplifier 500 includes a
polarization-maintaining (PM), large-mode area (LMA) fiber 532 that
is pumped at 976 nm with up to 400 W of power (e.g., supplied by a
laser-diode array, in some embodiments); the pump-delivery fiber
528 has a core diameter of 600 microns (in some embodiments, having
a numeric aperture of 0.22=NA). In some embodiments, LMA fiber 532
is double-clad, with a core diameter of 20 microns (in some
embodiments, having a numeric aperture of 0.06=NA) and an inner
cladding diameter of 400 microns (in some embodiments, having a
numeric aperture of 0.46=NA). In some embodiments, LMA fiber 532 is
wound on a water-cooled mandrel for cooling the LMA fiber 532. In
some embodiments, the seed beam 521 and pump beam 523 are both
coupled (in a direction left-to-right in FIG. 5A) via free-space
optics 530 into the input facet 531 of the LMA fiber 532, leaving
each output end 533 free to be spatially arrayed as required for
input to the multi-channel SBC (e.g., 430 of FIG. 4). In some
embodiments, the LMA fiber input facet 531 is held in a
water-cooled chuck that is mounted on a 5-axis, XYZ, tip/tilt
stage; both ends of the fiber 532 have 8-degree angle polishes. In
some embodiments, two free-space isolators 514 and 515 protect the
seed laser 550 from back-propagating signals generated by the power
amplifier 532. Table 1 provides a detailed list of components for
one embodiment of the system illustrated in FIG. 5A. TABLE-US-00001
TABLE 1 List of components illustrated in FIG. 5A. All lenses are
anti-reflection coated Component Vendor Part No. Description
Objective 1 ThorLabs C110TM-B 6.24 mm focal length molded glass
aspheric lens Objective 2 Special 54-15-15- 15 mm focal length,
fused Optics 980-1080 silica multi-element objective Objective 3
Special 54-17-30- 30 mm focal length, fused Optics 980-1080 silica,
multi-element objective Isolators Electro- 8I1055-WP Broadband
(1030-1080 nm), Optic free-space optical isolators, Technol- >30
dB isolation ogies Two isolators in series Dichroic Semrock
ACUL-0002 Beamsplitter that transmits beamsplitter 1040-1080 nm and
reflects 980 nm
[0146] FIG. 5B is a schematic plan view of a low-power optical
master oscillator system 550 for use in a MOPA-SBC system. In some
embodiments, the output is pulsed (5 ns pulses @ 10 MHz repetition
rate, in order to suppress SBS) and contains >300 mW of average
power with a spectral purity of >99.9% and a spectral bandwidth
of about 100 MHz. The wavelength is tunable over nearly the entire
Yb-gain spectrum. In some embodiments, the seed input 521 to the
LMA amplifier 532 of FIG. 5A is provided by the polarized,
low-power front-end laser depicted in FIG. 5B. In some embodiments,
the seed laser 550 includes a tunable, narrowband oscillator 560
(in some embodiments, a Littman-Metcalf external-cavity diode laser
(ECDL) having cavity reflectors 563 and 566, a laser tuning
mechanism 564 to tune the central wavelength, a laser diode 565, a
narrow bandpass tunable filter 567, and one or more serially
connected optical isolators 568), a Mach-Zehnder modulator 573, and
a series of single-mode PM gain stages. The ECDL output is
delivered to gain stage 1 via single-mode PM fiber through a PM-PM
splice 569. An isolator 568 is integrated into the ECDL housing to
protect the laser 560 from optical feedback. In some embodiments,
gain stage 1 includes a first pump laser diode 575 connected to one
input of multiplexer 570 through a PM-PM splice 569, where the
other input to multiplexer 570 through a PM-PM splice 569 is laser
560. The output of multiplexer 570 is connected to gain fiber 571
through a PM-Specialty Fiber splice 579, and the output of gain
fiber 571 is connected through a PM-Specialty Fiber splice 579 to a
1% tap 572 that extracts 1% of the signal to be used for
diagnostics and/or control (e.g., to control wavelength drift). The
output of the 1% tap is coupled through a PM-PM splice 569 to
Mach-Zehnder modulator 573, which converts the continuous-wave (cw)
optical output from stage 1 to a quasi-cw pulse train (5 ns pulses
@ 10 MHz repetition rate) that is designed to suppress Stimulated
Brillouin Scattering (SBS) in the LMA power amplifier 500 by
providing the narrowband seed on time scales that are short
compared to the about 10 ns SBS build-up time (see G. P. Agrawal,
Nonlinear Fiber Optics, 3rd ed. (Academic, New York, 2001)).
[0147] Subsequent amplification in gain stages 2 and 3 then
increases the average signal power to >300 mW. In some
embodiments, the output of modulator 573 is connected through a
PM-PM splice 569 to optical isolator 574, and then through a PM-PM
splice 569 to one input of multiplexer 577 through a PM-PM splice
569, where the other input to multiplexer 577 is connected to a
second pump laser diode 576 through a PM-PM splice 569. The output
of multiplexer 577 is connected to gain fiber 578 through a
PM-Specialty Fiber splice 579, and the output of gain fiber 578 is
connected through a PM-Specialty Fiber splice 579 to one input of
multiplexer 581, where the other input to multiplexer 581 is
connected to the output of a third pump laser diode 580 through a
PM-PM splice 569. The output of multiplexer 581 is connected to
gain fiber 582 through a PM-Specialty Fiber splice 579, and the
output of gain fiber 582 is connected through a PM-Specialty Fiber
splice 579 to output coupling 589, which couples the light to power
amplifier 500. The 1% tap coupler 572, inserted between gain stage
1 and the Mach-Zehnder modulator 574, taps a portion of the signal
to be used to monitor and/or control the ECDL fiber coupling and
the frequency stability of the ECDL output. Table 2 provides a
detailed list of the components used for some embodiments of the
front-end seed laser. TABLE-US-00002 Component Vendor Part No.
Description External Cavity Sacher SYS-500-1060-30 1055 nm center
wavelength Diode Laser Lasertechnik .+-.20 nm tuning range (ECDL)
559 30 mW power, CW, linear polarization ECDL fiber Sacher
FC-SMF-BCO Single-mode fiber coupling with anamorphic coupling
module Lasertechnik beam correction Integrated into ECDL housing
ECDL optical Sacher ISO-35-1060 35 dB optical isolator isolator 568
Lasertechnik Integrated into ECDL housing 980-nm 1.sup.st pump JDS
Uniphase 26-7602-180 180 mW fiber-coupled diode laser @ 976 nm
diode 575 980-nm 2.sup.nd pump JDS Uniphase 29-7602-400 400 mW,
fiber-coupled diode laser @ 976 nm diode 576 980-nm 3.sup.rd pump
JDS Uniphase 29-7602-500 500 mW, fiber-coupled diode laser @ 976 nm
diode 580 MUX 570, 577, 581 Novawave PMFWDM-9806-N-B-Q Polarization
maintaining wavelength division multiplexer, 980 pass/1060 reflect
1% Tap 572 Novawave PMFC-06-1-01-N-B-P-Q-F Polarization maintaining
1% tap coupler Modulator 573 The Cutting AM.10.03.H.PP.B.A
Polarization maintaining, Electro-optic Edge amplitude modulator
Isolator 574 Novawave PMI-1-06-P-N-B-Q 1064 nm polarization
maintaining optical isolator 1.sup.st Gain Stage 571 INO YB 500
Polarization maintaining, single-mode, Yb-doped optical fiber, 5
meters long 2.sup.nd Gain Stage 578 INO YB 500 Polarization
maintaining, single-mode, Yb-doped optical fiber, 10 meters long
3.sup.rd Gain Stage 582 INO YB 500 Polarization maintaining,
single-mode, Yb-doped optical fiber, 2 meters long
[0148] The wavelength tuning range for the front-end output is set
by the tuning range for the ECDL, which, in some embodiments,
provides single longitudinal mode output for wavelengths spanning
about 1040 nm to about 1080 nm (corresponding to a significant
fraction of the Yb gain bandwidth). This feature utilizes one of
the primary strengths for SBC, namely the ability to combine a
large number of channels spanning a broad range of wavelengths.
[0149] Broad range (>200 nm) spectral measurements of the
front-end seed-laser output for some embodiments indicate that
about 97.5% of the output power is contained in the narrowband
signal, with the remaining about 2.5% emitted as broadband
amplified spontaneous emission (ASE). To ensure a spectrally clean
seed for the LMA power amplifiers, in some embodiments, the
unwanted ASE component is removed with a custom interference filter
567 (in some embodiments, this provides 0.7 nm full-width at
half-maximum spectral bandwidth, such as a device available from
BARR Associates Inc, quotation no. 0507-1028CQ) placed immediately
after the front-end output. The center wavelength for the filter
can be angle tuned from less than 1055 nm to more than 1080 nm,
leads to negligible signal loss over this range, and improves the
front-end spectral purity to more than 99.9%. The resulting output
spectrum is shown in FIG. 6A.
[0150] FIG. 6A is a graph of an output spectrum of a seed laser
beam filtered using an angle-tuned filter.
[0151] FIG. 6B is a graph that shows high-resolution spectral
measurements of the front-end output obtained with a scanning
Fabry-Perot etalon with a 8 GHz free-spectral range and a finesse
of about 250 (corresponding to a spectral resolution of about 30
MHz). As shown in FIG. 6B, in some embodiments, the signal
linewidth is about 100 MHz.
[0152] FIG. 6C is a graph of measured values for one embodiment of
the LMA power amplifier's 1-micron-signal output versus power
exiting the pump-delivery fiber. In all cases, the pump wavelength
is temperature-tuned to about 976 nm. Key points from the data
include: [0153] (1) the amplifier's slope efficiency with respect
to the launched pump power is about 73% (i.e., 52% with respect to
power from pump fiber), and [0154] (2) the maximum output power of
208 W is limited by the available pump power. Measurements of the
amplified signal PER and polarization stability with respect to
time for amplified signals in excess of 100 W indicate the PER is
greater than 20 dB and the polarization is stable at the level of
.+-.2% over time scales of several minutes. Moreover, again for
amplified signals in excess of 100 W, the measured amplified signal
spectral purity is greater than 99%, and the
full-width-half-maximum spectral bandwidth is less than 1 GHz.
[0155] The beam quality for the 1-micron signal (i.e., a signal
having wavelengths of about 1000 nanometers, for example, a signal
having wavelengths in the range of about 1000 nm to about 1100 nm,
in some embodiments) at the LMA amplifier's output, at signal
powers relevant to high power SBC, has also been characterized for
some embodiments. For these measurements, the 1-micron signal is
first collimated with a multi-element objective and then passed
through a fused silica wedge. The first-surface reflection from the
wedge is then focused to a waist with an f=10-cm singlet lens and
M.sup.2 for the amplified signal is determined via the knife-edge
technique.
[0156] FIG. 6D is a graph of representative M.sup.2 measurements
along the horizontal-direction transverse beam axis of the 1-micron
signal, collected at an amplified average power of 100 W.
[0157] FIG. 6E is a graph of representative M.sup.2 measurements
along the vertical-direction transverse beam axis of the same
1-micron signal. Open circles in FIGS. 6D and 6E give measured
1/e.sup.2 beam radii, while solid lines in FIGS. 6D and 6E are the
algebraic fits to the standard expression: .omega. .function. ( z )
= .omega. 0 .times. 1 + ( M 2 .times. .lamda. .function. ( z - z 0
) .pi. .times. .times. .omega. 0 2 ) 2 EQ . .times. ( 13 )
##EQU15## where .omega.(z) is the 1/e.sup.2 beam radius (the "beam
waist") at position z, .omega..sub.0 is the 1/e.sup.2 beam radius
at position z.sub.0, and .lamda. is the laser wavelength. Fitting
this expression to the data in FIGS. 6D and 6E gives M.sup.2=1.06
for both the horizontal and vertical beam axes, clearly indicating
nearly diffraction-limited beam propagation.
[0158] As a first step toward demonstrating the ability to obtain
greater than 100 kW of high-quality 1-micron light via
spectral-beam combining of outputs of multiple Yb fiber lasers,
some embodiments have used MOPAs with the characteristics outlined
above to demonstrate: [0159] (1) two-channel fiber laser SBC with a
power-combining efficiency of 93%, a combined beam power of 258 W,
and a dispersed beam axis M.sup.2 of 1.06, and [0160] (2)
three-channel fiber laser SBC with a power-combining efficiency of
93%, a combined beam power of 90 W, and a dispersed axis M.sup.2 of
1.01.
[0161] FIG. 6I is a schematic diagram of the optical layout
utilized for the two-channel demonstration. The optical layout
utilized for the three-channel demonstration is depicted in FIG.
6J. In both cases, SBC is performed with a method more akin to
wavelength-division multiplexing (such as shown in FIG. 3A and
described above) than standard spectral beam combining (see S. J.
Augst, A. K. Goyal, R. L. Aggarwal, T. Y. Fan and A. Sanchez,
"Wavelength beam combining of ytterbium fiber lasers," Opt. Lett.,
vol. 28, 331-333 (2003), and T. Y. Fan, "Laser Beam Combining for
High-Power, High Radiance Sources," IEEE Journal of Selected Topics
in Quantum Electronics, vol.11, 567-577 (2005)) but where the MOPA
wavelengths used as inputs for FIG. 6I and FIG. 6J are individually
controlled (as shown in FIG. 5B, FIG. 7A or FIG. 7C) rather than
set with a common external cavity (as shown in FIG. 3A, FIG. 3D) or
a commonly applied fiber grating formation (as shown in FIG. 3F).
As shown in FIG. 6I and FIG. 6J, in some embodiments, the
individual MOPA outputs are separately collimated with f=30 mm
multi-element objectives (Special Optics Part No.54-17-30-980-1080)
and then directed via free-space mirrors (CVI Laser Optics Part No.
Y1-1037-45-P-UV) to a multi-layer dielectric diffraction grating
(in some embodiments, made by methods described in, e.g., M. D.
Perry et al., "High-efficiency multilayer dielectric diffraction
gratings," Opt. Lett., vol. 20, 940-942 (1995) (and U.S. Pat. No.
5,907,436 entitled "Multilayer dielectric diffraction gratings"
issued May 25, 1999 to Perry et al.); and Karl Hehl et al.,
"High-efficiency dielectric reflection gratings: design,
fabrication, and analysis," Appl. Opt., vol. 38, 6257-6271 (1999))
with a groove density of 1740/mm. Images acquired with a
charge-coupled device (CCD) camera are then used to overlap the
beams both on the grating and in the far field.
[0162] For the two-channel demonstration, the grating was aligned
for Littrow reflection at 1065 nm in the horizontal plane (the
grating-dispersion plane) and tipped by 1.7 degrees in the vertical
direction. For the three-channel demonstration, the grating was
aligned for Littrow reflection at 1060 nm in the horizontal plane
and tipped by 0 degrees (i.e., not tipped) in the vertical
direction. To ensure optimum first-order diffraction efficiency,
zero-order half-wave plates (CVI Laser Optics part no.
QPM-1064-10-2) are used, in some embodiments, to orient the
separate beam polarizations on the grating. The measured
first-order efficiency at all three wavelengths is about 94%. The
SBC power-combining efficiency, defined as the combined beam power
divided by the sum of the individual MOPA output powers, is about
93%. The combined beam then passes between the input beams and onto
an optical power meter. Roughly 1% of the two-channel (or 5% of the
three-channel) combined beam power incident on the power meter is
directed, with a fused silica wedge, to an optical spectrum
analyzer (OSA), a beam propagation analyzer, and the CCD camera.
The OSA is used to determine the spectral content of the combined
beam over a broad spectral range while the beam propagation
analyzer is used to determine the combined beam quality along both
the dispersed and non-dispersed beam axes.
[0163] Obtaining high-quality SBC-combined beams requires the
dispersive beam-combining element to remain nearly distortion free
when irradiated by the full output power from the fiber MOPAs. To
test the power-handling capabilities of the diffraction grating,
some embodiments have performed interferometric measurements of the
grating surface distortion induced by high irradiance optical
loads.
[0164] FIG. 6F is a schematic diagram of the basic test setup to
test the power-handling capabilities of a diffraction grating. The
grating is inserted into the sample arm of a Michelson
interferometer probed by a 633 nm beam and aligned for Littrow
reflection at 633 nm; hence the quality of the diffracted order is
probed directly. The optical load, or heating beam, is provided by
a 1080 nm Yb fiber MOPA. The heating beam's polarization is aligned
for maximum first-order diffraction efficiency.
[0165] FIG. 6G shows an interferogram recorded when the heating
beam is off.
[0166] FIG. 6H shows an interferogram recorded when the heating
beam peak irradiance on the grating surface is 1.5 kW/cm.sup.2. The
latter case employed 70 W of 1080-nm power. The image in FIG. 6H
indicates that 1.5 kW/cm.sup.2 optical loads generate about
0.2-wave distortion at 633 nm; roughly equivalent to 0.1-wave
distortion at 1-micron wavelengths. This level of distortion is
expected to have a negligible impact on the SBC-combined beam
quality.
[0167] FIG. 6K is a graph of an optical spectrum for the 175-W
two-channel SBC-combined beam with 93 W at 1065-nm wavelength and
82 W at 1055-nm wavelength. The signal-to-noise ratio for the
combined beam is greater than 50 dB, clearly indicating that
optical power outside the signal wavelengths (due, for example, to
amplified spontaneous emission) makes a negligible contribution to
the combined beam power.
[0168] FIG. 6L is a graph of an optical spectrum for the 90-W
three-channel SBC-combined beam (with about 30 W each at 1055 nm,
1062.5 nm, and 1065 nm). The combined beam signal-to-noise is
greater than 50 dB, again clearly indicating that optical power
outside the signal wavelengths makes a negligible contribution to
the combined beam power. (The high signal-to-noise ratio is
observed for all power levels observed to date, including the
maximum two-channel value of 258 W for some embodiments.) The
measurements for FIG. 6K and FIG. 6L were performed with a
single-mode fiber-coupled OSA (having 0.2 nm resolution
bandwidth).
[0169] FIG. 6M is a graph of M.sup.2 measurements along the
dispersed direction transverse beam axis for a 175-W two-channel
SBC-combined beam (93 W at 1065 nm, 82 W at 1055 nm). FIG. 6N is a
graph of M.sup.2 measurements along the non-dispersed direction
transverse beam axis for the same combined beam. Measurements were
performed with a Coherent Modemaster beam profiler. For the two
channel demonstration, the peak intensity on the diffraction
grating is about 450 W/cm.sup.2.
[0170] FIG. 6O is a graph of M.sup.2 measurements along the
dispersed direction transverse beam axis for the 90-W three-channel
SBC-combined beam (roughly 30 W each at 1055 nm, 1062.5 nm, and
1065 nm). FIG. 6P is a graph of M.sup.2 measurements along the
non-dispersed direction transverse beam axis for the same combined
beam. Measurements here were also performed with a Coherent
Modemaster beam profiler. For the three-channel demonstration, the
peak intensity on the diffraction grating is about 230 W/cm.sup.2.
For FIG. 6M, FIG. 6N, FIG. 6O and FIG. 6P, the open circles give
measured 1/e.sup.2 beam radii while the solid lines are fits to EQ.
13.
[0171] In both cases (the 175-W two-channel SBC-combined beam and
the 90-W three-channel SBC-combined beam), the beam quality along
the dispersed and non-dispersed beam axes is essentially identical
to the beam quality for the direct output from a single-fiber MOPA.
This result clearly indicates that the combined beam quality is not
significantly influenced by either the spectral bandwidth for the
individual MOPAs or surface distortion of the grating.
[0172] FIG. 6Q is a graph of the dispersed-axis M.sup.2 for the
two-channel SBC-combined beam versus combined beam power. In all
cases, the combined beam contains roughly equal amounts of power at
1055 nm and 1065 nm. As shown by FIG. 6Q, this trend continues all
the way to combined beam powers of 258 W, where each MOPA
contributes about 130 W to the combined beam and the peak intensity
on the grating is about 650 W/cm.sup.2. Moreover, the FIG. 6Q data
are essentially flat with respect to the combined beam power, a
result that strongly suggests combined beam powers in excess of 260
W can be obtained with similarly excellent beam quality.
[0173] FIG. 7A is a schematic plan view of a spectral-beam-combiner
laser system 700 with wavelength-dispersion compensation. In some
embodiments, system 700 includes a plurality of MOPA fiber lasers,
each having a master oscillator seed laser 550 tuned to a different
wavelength followed by its own fiber power amplifier 500. The
laser-beam outputs of the plurality of MOPA lasers are combined
into a single beam using a spectral beam combiner 130, as described
above. In some embodiments, each laser is pulsed on a schedule
controlled by pulse-timing circuit 781, while in other embodiments,
continuous-wave (CW) lasing is used. In some embodiments, the pump
lasers are supplied by one or more pump-laser-diode power supplies
794 and 782. In some embodiments, the fibers of the power
amplifiers are wound around a water-cooled mandrel cooled by
fiber-cooling water supply 783. In other embodiments, other
suitable cooling mechanisms are used, such as air cooling, or
refrigerant cooling.
[0174] Note that, in general, if the two gratings of the SBC are
kept parallel, then each component of the output beam 90' will be
parallel to its respective input beam from one of the power
amplifiers 500, and thus as a laser's wavelength drifts (i.e.,
changes over time due to, e.g., temperature changes of the
linewidth-narrowing filter) the portion of output beam 90' due to
that laser, while remaining parallel to the main output beam 90',
will move off center of main output beam 90'. In some embodiments,
a real-time diagnostic-and-adjustment unit 709 is used to
dynamically adjust the wavelengths of the individual lasers in
order that every laser's beam is centered in the single output beam
90'.
[0175] In some embodiments, beam centering is accomplished by
detecting whether the particular beam is parallel but not aligned
(i.e., the beam does not hit the single spot on the second
diffraction grating to which the other beams are directed) into the
single output beam 90', for example, using a detector 711 (e.g.,
that receives light only if a beam is too high) and a detector 712
(e.g., that receives light only if a beam is too low), both of
which are connected to off-axis circuit 732. In some embodiments,
off-axis circuit 732 includes an output 736 that is connected to
and analyzed by real-time wavelength-drift diagnostic circuit 786,
whose output 735 controls individual wavelength adjustment circuits
787 (e.g., in some embodiments, these control, for example, the
resonant wavelength of the initial seed laser or its output filter)
for each laser output. In some embodiments, circuit 786 has one or
more inputs 784 that are connected to receive pulse timing
information (e.g., from pulse timing circuit 785), in order to
determine which laser needs its wavelength adjusted. In some
embodiments, circuit 786 has outputs (not shown) that are connected
to transmit pulse timing information (e.g., to pulse timing circuit
785), in order to control pulse timing and/or laser power, in order
to determine which laser needs its wavelength adjusted.
[0176] Thus, if the wavelength of one of the seed lasers drifts,
its output, while remaining parallel, will also drift off center,
and real-time diagnostic-and-adjustment unit 709 detects which of
the lasers is off-center and automatically adjusts its wavelength
until its portion of the output beam is again centered. In some
embodiments, off-axis circuit 732 optionally includes an output 734
that controls individual positioners (e.g., in some embodiments,
five-degrees-of-freedom positioners that control, for example, X,
Y, Z, pitch angle, and yaw angle) for each laser's output (the
input to SBC 130). In some embodiments, a combination of wavelength
control and positioning control is used to keep all beams parallel
and aligned into the single output beam 90' by iteratively
adjusting angle (using five-degrees-of-freedom positioners) and
position (using laser wavelength and/or the five-degrees-of-freedom
positioners) on each beam.
[0177] Note that, if the two gratings of the SBC are kept parallel,
then each component of the output beam 90' will be parallel to its
respective input beam from one of the power amplifiers 500, so if
one of the input beams is not parallel (i.e., that beam strikes the
first diffraction grating at an angle different than the angle of
the other input beams), its portion of the output beam will also
not be parallel and will diverge at an angle from the main output
beam 90' corresponding to the angle error of the input beam and a
correction will be needed. In some embodiments, this is
accomplished by detecting whether the particular beam is initially
aligned into the single spot on the second grating, but is angled
too high, for example, using detector 713 (e.g., that receives
light only if a beam is angled too high) and detector 714 (e.g.,
that receives light only if a beam is angled too low), both of
which are connected to off-angle (angle positioning) detector and
diagnostic circuit 731, which detects which input laser beam is
off-angle, and whose output 733 controls the individual positioners
(e.g., in some embodiments, five-degrees-of-freedom positioners
that control, for example, X, Y, Z, pitch angle, and yaw angle) for
each laser's output (the input to SBC 130). In some embodiments,
off-angle (angle positioning) detector and diagnostic circuit 731
also includes an output signal 737 that is used in combination with
the other inputs by diagnostic circuit 786 to control the
wavelengths of the seed lasers. In some embodiments, circuit 731
has inputs (not shown) that are connected to receive pulse timing
information (e.g., from pulse timing circuit 785), in order to
determine which laser needs its angle adjusted. In some
embodiments, circuit 731 has outputs (not shown) that are connected
to transmit pulse timing information (e.g., to pulse timing circuit
785), in order to control pulse timing and/or laser power, in order
to determine which laser needs its angle adjusted.
[0178] In some embodiments, the other beams (those not being
adjusted) can be fully on (either CW or at their normal pulse
schedule) while the adjustments are made to the particular beam of
interest. In some embodiments, the particular beam of interest is
turned on or off, or its power is increased or decreased, and the
detection circuits 731 and/or 732 detect whether a corresponding
change in the off-axis or off-angle signal is observed. This is a
major advantage of the present invention for those circumstances
where it is desired to adjust the laser system's parameters (i.e.,
the wavelength of one or more beams) while keeping all of the other
beams operating and thus maintaining nearly full output power of
the combined beam 90'. The flowchart of one such process according
to the present invention is shown in FIG. 7B.
[0179] FIG. 7B is a flowchart of a real-time diagnostic and
adjustment process 770 according to some embodiments of the
invention. In some embodiments, the process starts at block 771,
and at block 772, the process detects that one or more beams is
off-axis and/or off-angle. At block 773, one of the plurality of
lasers is selected, and that laser is turned on or off, and/or its
power is increased or decreased, and at decision-block 774 the
process either: if no change was detected in the off-axis and/or
off-angle signal, then branches back to block 773 to select the
next laser and that laser is turned on or off, and/or its power is
increased or decreased; or if a corresponding change was detected
in the off-axis and/or off-angle signal, then branches to block
775, where the wavelength and/or position of the selected laser is
adjusted and control passes to block 776 and this laser is again
turned on or off, and/or its power is increased or decreased and
control passes again to decision block 774 to iterate until the
selected laser is adjusted and aligned. This entire process is
repeated for every laser beam. If process 770 detects that two or
more beams are off-axis and/or off-angle (e.g., if the on-off
action is performed for each single laser and the off-axis and/or
off-angle signal does not detect a change (e.g., the off-axis
and/or off-angle signal stays on of two lasers are off-axis and/or
off-angle, since when one laser is cycled on and off, the other
laser remains on and saturates the detectors 711-714), then the
process turns off one or more of the other lasers while testing the
particular selected laser until a single off-axis and/or off-angle
beam can be adjusted by itself, whereupon one of the other lasers
can be turned back on and the adjustment process continued. In some
embodiments, the process of FIG. 7B is performed for any of the
embodiments described herein, including (but not limited to) those
of FIG. 7A and FIG. 7C.
[0180] FIG. 7C is a schematic plan view of a spectral-beam-combiner
laser system 701 with wavelength-dispersion compensation. System
701 is substantially similar to system 700 of FIG. 7A, except that
a separate off-wavelength detector/adjuster 708 is used on laser
beams of a plurality of the master oscillators 550 before the power
amplifiers, in order that the wavelengths of the master oscillators
550 (the seed lasers) can be continuously monitored and adjusted
without or before turning on the power amplifiers, in order that
when the power amplifiers are turned on, all of the wavelengths
will be stable and aligned to the desired wavelengths. In some
embodiments, each laser is activated during only one pulse of a
stream of pulses (e.g., if one hundred lasers each having a
different wavelength are provided, and each laser is active for a
1/100 duty cycle that is staggered with the other lasers, then a
substantially continuous output laser beam can be provided). For
example, if a first laser (A) is activated during the third pulse
722 of a pulse stream 721 and its pulse corresponds to a pulse
detected by lower detector 712, then its wavelength will be
adjusted (e.g., by shortening the wavelength) until its beam is
again aligned with the main beam and not detected by lower detector
712. Similarly, for example, if a second laser (B) is activated
during the ninth pulse 723 of pulse stream 721 and its pulse
corresponds to a pulse detected by upper detector 711, then its
wavelength will be adjusted (e.g., by lengthening the wavelength)
until its beam is again aligned with the main beam and not detected
by upper detector 711. In some embodiments, circuit 701 also
includes off-angle and/or off-axis detectors and diagnostic
circuits associated with its output beam 90' as shown in FIG.
7A.
[0181] The following description is more completely described in
U.S. Provisional Patent Application 60/703,824 filed on Jul. 29,
2005, titled "PERIODIC FIBER TO SUPPRESS NONLINEAR EFFECTS IN
RARE-EARTH-DOPED FIBER AMPLIFIERS AND LASERS" which is hereby
incorporated by reference in its entirety, wherein power-amplifier
fibers having low-NA large mode area (LMA) straight sections are
connected using high-NA corner sections of fiber, in order to
obtain a more compact high-power fiber amplifier.
[0182] Another element of the invention involves the use of
dielectric-coated gratings as one or more of the dispersive
beam-combining elements in the system. Common gratings are
typically coated with a metallic reflective surface, for example,
gold or aluminum. Such materials absorb a few percent of the light
impinging upon them. This causes heating of the grating surface,
which can lead to optical distortions, and at sufficiently high
power levels, to optical damage of the grating. Recently, progress
has been made in grating technology, which allows gratings with
coatings related to those used in dielectric mirrors to be
produced. These gratings allow the beam-combined fiber laser system
to operate at very high power levels.
[0183] Further, some embodiments provide diagnostic circuits and/or
processes to properly set the operating wavelengths of the various
fiber oscillator/amplifiers in the system so their beams will be
overlapped in the far field, producing a single high-quality,
focusable beam. By utilizing a time-gated beam position diagnostic
synchronized to the pulsing of one or more of the laser
oscillators, and/or staggering the firing of the laser oscillators,
the present invention easily differentiates among the laser
oscillators and thus provides control signals to the correct
oscillator as selected by the diagnostics.
[0184] In general, fiber nonlinear effects are directly
proportional to the effective power intensity in the core and
inversely proportional to effective fiber length. Either increasing
fiber core diameter or reducing fiber length in a fiber
amplifier/laser will mitigate nonlinear effects. To maximize power
handling of the fiber amplifier/laser, the large core section can
have a core diameter tens of times larger than that of a
conventional LMA (large mode area) fiber, but still provide very
good beam quality if the NA (numerical aperture) is kept low
enough. Furthermore, increasing core diameter not only reduces
signal power intensity (thus reducing non-linear effects), but also
increases the fiber's pump absorption if the pump cladding outer
diameter is unchanged along the length (since the pump light
confined and bouncing around in the inner cladding is more likely
to intercept the larger core). As a result, in some embodiments, an
increase in the core diameter by a factor of two increases the
threshold of nonlinear effects by a factor of sixteen. This is
because the area of the fiber core increases by a factor of four,
and the fiber length can be decreased by a factor of four due to
increased pump absorption by the large core.
[0185] In some embodiments, the optical-fiber power amplifiers
include a plurality of large core area (and thus LMA), low NA
straight sections serially connected to one another by short,
small-core, high NA curved sections to form a "periodic" fiber. In
some embodiments, the small-diameter core sections are much shorter
than the large-diameter core sections, so nonlinear effects do not
have enough gain to reach threshold in the small core sections.
Furthermore, the unique structure of the fiber will also reduce
fiber nonlinear effects through multiple mechanisms. For example,
the change of core NA broadens the SBS gain spectrum due to
nonhomogeneous broadening of SBS so that the SBS threshold is
increased. In some embodiments, the change of core diameter may
also affect fiber chromatic dispersion so that it increases
thresholds of some nonlinear effects, which require phase-matching,
such as four-wave mixing.
[0186] The unique structure will also simplify system configuration
for using mode-matching techniques to achieve good beam quality for
slightly multimode fiber amplifiers. For a rare-earth (RE)-doped
multimode fiber, the signal launch condition plays an important
role in mode selection, as described in U.S. Pat. No. 5,818,630
(incorporated herein by reference). Even though the RE-doped core
supports several different modes, the high beam quality of a
single-mode source can be preserved in the multimode amplifier if
the mode-field-diameter (MFD) of the source is matched with that of
the fundamental mode of the multimode fiber. As a result, a
RE-doped multimode fiber amplifier can achieve diffraction-limited
beam quality. An elegant method for achieving such mode matching in
the fiber structure of the present invention involves designing the
small core section to have the same core diameter and NA as that of
the signal fiber from the seed source. Then, by cleaving or
otherwise cutting the fiber in the small-core section (and
preparing the fiber end by techniques well-known to those skilled
in the art) and injecting the mode to be amplified, the required
mode-matching is achieved by the hybrid fiber itself. There is no
need to use additional components to match MFDs between the signal
fiber and the amplifier fiber.
[0187] In some embodiments, the periodic fiber can be bent at the
small-core sections, enabling it to be packaged in a compact
enclosure. Unlike a conventional LMA fiber, a very large core,
extremely low-NA uniform fiber must be kept straight, or be coiled
at a large radius, whereas the periodic fiber structure described
here has much more flexibility in terms of packaging. The
small-core sections can be designed to be very short with a small
core diameter and large NA, so that they can be bent in a small
diameter with negligible bending loss. The locations of the
small-core sections can be periodic for packaging simplicity, or
with a varied period to meet a certain packaging requirement.
[0188] FIG. 8A is a schematic plan view of a ribbon-fiber MOPA
spectral-beam-combiner laser system 800. System 800 is
substantially identical to system 306 if FIG. 3F described above,
but with the substitution of a one or more ribbon fibers used for
amplifiers 855 and 860 and laser active media 851 and 856 in order
to obtain very close center-to-center spacings of the core ends
(for example, in some embodiments, ribbon fibers such as described
in the paper by L. J. Cooper et al., High-power Yb-doped multicore
ribbon fiber laser, Optics Letters Vol. 30, No. 21, Nov. 1, 2005,
which is incorporated herein by reference), each having a plurality
of doped cores within an inner core used to carry pump light and
inject it into the cores (e.g., Yb doped silica cores arranged
side-by-side along a straight cross-section line within a
substantially pure silica inner cladding having a width
substantially larger than its height, and a polymer outer cladding
for keeping the pump light contained). In some embodiments, each
power amplifier core in fiber ribbons 855 and 860 is a
large-diameter (e.g., about 40 to 50 microns or larger) designed to
have a low NA large mode area and operated in substantially only
its lowest-order mode (LP01). In some embodiments, a
photonic-crystal configuration is used to define the boundaries of
the cores, in order to obtain a well-controlled very low NA for
each core, for example, as described for FIG. 8B below. In other
embodiments, individually formed inner-cladding fibers, each having
one core, are held within a single outer polymer cladding in a
similar configuration. In still other embodiments, each power
amplifier fiber, rod, rod-like fiber, or ribbon fiber has a portion
of its outer section shaved or ground away, in order to obtain very
close center-to-center spacings of the core ends.
[0189] FIG. 8B is a schematic cross-section view of a
photonic-crystal (PC) ribbon-fiber 851. In some embodiments, PC
ribbon-fiber 851 has a plurality of active amplifying cores 856
(e.g., Yb-doped silica or aluminosilicate) within an inner cladding
854 (e.g., silica) surrounded by outer cladding 853 (e.g.,
polymer); each core 856 surrounded by a plurality of
longitudinally-oriented holes (e.g., air-filled) or similarly
shaped regions of lower index of refraction, in order to well
define a very low NA core in order to support a single lowest-order
mode (LP01). In some embodiments, the cores, for example, are each
40 to 50 microns diameter or larger with a center-to-center
spacing, in some embodiments, of 100 microns or less within a
ribbon approximately 250 microns high and 1 mm or more wide.
[0190] Some embodiments of the present invention provide a method
that includes providing a first laser beam and a second laser beam,
introducing a first chromatic dispersion into the first laser beam,
introducing a second chromatic dispersion into the second laser
beam, introducing a third chromatic dispersion into the first laser
beam, wherein the third chromatic dispersion compensates for the
first chromatic dispersion, introducing a fourth chromatic
dispersion into the second laser beam, wherein the fourth chromatic
dispersion compensates for the second chromatic dispersion, and
combining the first and second laser beams into a single output
beam. In some embodiments, the first compensation includes
eliminating substantially all further chromatic dispersion from the
first laser beam, and the second compensation includes eliminating
substantially all further chromatic dispersion from the second
laser beam. In some embodiments, the first and second chromatic
dispersions are generated by a first diffraction grating and the
compensating third and fourth chromatic dispersions are generated
by a second diffraction grating, wherein, for each respective
diffracted beam, the angle at which each beam leaves the first
grating surface is equal to the angle at which that beam approaches
the second grating surface.
[0191] Some embodiments further include providing a first
diffractive element, wherein the introducing of the first chromatic
dispersion includes diffracting the first laser beam with the first
diffractive element, and the introducing of the second chromatic
dispersion includes diffracting the second laser beam with the
first diffractive element, and providing a second diffractive
element, wherein the introducing of the third chromatic dispersion
includes diffracting the first laser beam with the second
diffractive element, and the introducing of the fourth chromatic
dispersion includes diffracting the second laser beam with the
second diffractive element, introducing the compensating chromatic
dispersions to each respective one of the plurality of diffracted
laser beams. In some embodiments, the first and second diffractive
elements are separate diffractive-reflection gratings. In some
embodiments, these are dielectric gratings wherein the diffractive
surface is formed on a top surface of a stack of dielectric layers
formed to provide a highly efficient diffractive reflection. In
some embodiments, these diffractive gratings are oriented at or
near a Littrow angle wherein the diffracted beam is at or near the
input beam but in an opposite direction.
[0192] In some embodiments, the providing of the first diffractive
element and the second diffractive element includes having the same
diffractive pattern on both the first diffractive element and the
second diffractive element. In some embodiments, the diffractive
pattern is about 1740 lines/mm in a direction perpendicular to the
input beams. In other embodiments, 1400 lines/mm or 1200 lines/mm,
or other suitable line densities are used.
[0193] Some embodiments further include positioning of the second
diffractive element so a diffractive surface of the second
diffractive element is parallel to a corresponding diffractive
surface of the first diffractive element. In some embodiments, the
first and second diffractive elements are separate
diffractive-reflection planar gratings oriented parallel to one
another. In other embodiments, the first and second diffractive
elements are separate areas on a single planar
diffractive-reflection grating, and, after the plurality of input
beams are first diffracted from the single diffractive-reflection
grating, one or more reflective surfaces are used to redirect the
beams to again diffract from the single diffractive-reflection
grating into a single combined beam.
[0194] Some embodiments further include providing a plurality of
Yb-doped singly clad or multiply clad optical fibers including a
first fiber and a second fiber, pumping an inner cladding of each
of the plurality of fibers with pump light from one or more laser
diodes, generating with the first fiber the first laser beam at a
first wavelength, and generating with the second fiber the second
laser beam at a second wavelength. In some embodiments, a
master-oscillator-power-amplifier (MOPA) configuration is used,
wherein a plurality of master oscillators are used to each generate
a very narrowband (e.g., in some embodiments, each about 10 GHz or
less FWHM linewidth) laser seed signal at each of a plurality of
different center wavelengths, and each of these narrowband signals
is amplified by its respective power amplifier, which includes one
or more serial amplification stages, in order to provide a
plurality of high-power laser beams that are to be spectrally
combined. In some embodiments, each power amplifier includes a
plurality of fiber-amplification stages, each stage separately
pumped, wherein each stage is separated by a narrowband wavelength
filter to prevent amplified stimulated emission (ASE). In some
embodiments, each laser seed signal is gated or otherwise
controlled to a pulse length sufficiently short to substantially
prevent stimulated Brillouin scattering (SBS) buildup. In some
embodiments, the pulse length is controlled to be about 10 ns or
less. In some embodiments, the pulse length is controlled to be
about 9 ns or less. In some embodiments, the pulse length is
controlled to be about 8 ns or less. In some embodiments, the pulse
length is controlled to be about 7 ns or less. In some embodiments,
the pulse length is controlled to be about 6 ns or less. In some
embodiments, the pulse length is controlled to be about 5 ns or
less. In some embodiments, the pulse length is controlled to be
about 4 ns or less. In some embodiments, the pulse length is
controlled to be about 3 ns or less. In some embodiments, the pulse
length is controlled to be about 2 ns or less. In some embodiments,
the pulse length is controlled to be about 1 ns or less. In some
embodiments, the pulse length is controlled to be about 0.5 ns or
less. In some embodiments, the pulse length is controlled to be
about 0.2 ns or less. In some embodiments, the pulse length is
controlled to be about 0.1 ns or less. In some embodiments, the
pulse length is controlled to be about 0.05 ns or less.
[0195] In some embodiments, as the pulse length becomes
sufficiently short, the spectral width of the pulse increases
(e.g., due to Fourier expansion of the short pulse length). In some
embodiments, a two-or-more grating spectral beam combiner (for
example, such as shown in FIG. 3D or FIG. 3G) is used to reduce or
eliminate further spectral spreading after the beams are
combined.
[0196] Some embodiments include filtering the first laser beam to a
full-width half-maximum linewidth of less than about one nanometer,
filtering the second laser beam to a full-width half-maximum
linewidth of less than about one nanometer, pulsing the first laser
beam to a pulse length of less than about ten nanoseconds and
sufficiently short to substantially prevent SBS buildup, and
pulsing the second laser beam to a pulse length of less than about
ten nanoseconds and sufficiently short to substantially prevent SBS
buildup. Some embodiments of the method further include tuning the
first fiber to generate the first laser beam at the first
wavelength, tuning the second fiber to generate the second laser
beam at the second wavelength, detecting that one of the laser
beams has become misaligned relative to the single combined beam,
determining that the first laser beam is the misaligned one, and
adjusting the tuning of the first fiber in order that the first
laser beam is aligned relative to the single combined beam.
[0197] In some embodiments, the detecting of the misaligned beam
and the determining that the first laser beam is the misaligned one
are based on a timing of a pulse of one of the laser beams.
[0198] In some embodiments, the detecting of the misaligned beam
and the determining that the first laser beam is the misaligned one
are performed while the second beam is active.
[0199] In some embodiments, the determining that the first laser
beam is the misaligned one is performed during a time when a
plurality of the other laser beams are "on" (i.e., actively
propagating laser beams), and includes: changing a power value of
the first laser beam, detecting a corresponding change in a
misaligned beam, and changing a tuning of the first laser based on
the detecting of the corresponding change in the misaligned
beam.
[0200] Some embodiments of the present invention provide a method
that includes providing a first diffractive element and a second
diffractive element, directing a plurality of light beams to a
plurality of locations on the first diffractive element, and
positioning the second diffractive element relative to the first
diffractive element such that the plurality of light beams
diffracted from the plurality of locations on the first diffractive
element are directed to a single location on the second diffractive
element and are diffracted by the second element into a single
combined beam.
[0201] Some embodiments of this method further include, with the
first diffractive element, introducing a chromatic dispersion to
each of the plurality of diffracted light beams, and with the
second diffractive element, introducing a compensating chromatic
dispersion to each of the plurality of diffracted light beams.
[0202] In some embodiments, the providing of the first diffractive
element and the second diffractive element includes having the same
diffractive pattern on both the first diffractive element and the
second diffractive element.
[0203] In some embodiments, the positioning of the second
diffractive element includes positioning the second diffractive
element so a diffractive surface of the second diffractive element
is parallel to a corresponding diffractive surface of the first
diffractive element.
[0204] Some embodiments further include generating a plurality of
laser beams for use as the plurality of light beams.
[0205] Some embodiments further include providing a plurality of
Yb-doped multiply clad optical fibers including a first fiber and a
second fiber, pumping an inner cladding of the plurality of fibers
with pump light from one or more laser diodes, generating with the
first fiber a first laser beam at a first wavelength for one of the
plurality of light beams, and generating with the second fiber a
second laser beam at a second wavelength for another of the
plurality of light beams.
[0206] Some embodiments further include tuning the first fiber to
generate the first laser beam at the first wavelength, tuning the
second fiber to generate the second laser beam at the second
wavelength, detecting that one of the laser beams has become
misaligned relative to the single combined beam, determining that
the first laser beam is the misaligned one, and adjusting the
tuning of the first fiber in order that the first laser beam is
aligned relative to the single combined beam.
[0207] In some embodiments, the detecting of the misaligned beam
and the determining that the first laser beam is the misaligned one
are based on a timing of a pulse of one of the laser beams.
[0208] In some embodiments, the detecting of the misaligned beam
and the determining that the first laser beam is the misaligned one
are performed while the second beam is active.
[0209] In some embodiments, the determining that the first laser
beam is the misaligned one is performed while a plurality of the
other laser beams are ON, and includes: changing a power value of
the first laser beam, detecting a corresponding change in a
misaligned beam, and changing a tuning of the first laser based on
the detecting of the corresponding change in the misaligned
beam.
[0210] Some embodiments of the present invention provide an
apparatus that includes a first diffractive element and a second
diffractive element, and a source of a plurality of light beams
directed to plurality of locations on the first diffractive
element, wherein the second diffractive element is positioned
relative to the first diffractive element such that the plurality
of light beams diffracted from the plurality of locations on the
first diffractive element are directed to a single location on the
second diffractive element and are diffracted by the second element
into a single combined beam. In some embodiments, the first and
second diffractive elements are separate diffractive gratings. In
some embodiments, these are dielectric gratings wherein the
diffractive surface is formed on a top surface of a stack of
dielectric layers formed to provide a highly efficient diffractive
reflection. In some embodiments, these diffractive gratings are
oriented at or near a Littrow angle wherein the diffracted beam is
at or near the input beam but in an opposite direction. In other
embodiments, the first and second diffractive elements are areas on
a single diffraction grating, and optical elements are positioned
such that beams leaving the grating after the first diffraction are
directed to again approach the grating for the compensating
diffraction that removes further chromatic dispersion.
[0211] In some embodiments, the first diffractive element
introduces a chromatic dispersion to each of the plurality of
diffracted light beams, and the second diffractive element
introduces a compensating chromatic dispersion (one of
substantially equal and opposite diffractive amount, in order to
stop further chromatic dispersion, and which combines the beams
into a single output beam) to each of the plurality of diffracted
light beams such that the single combined output beam has
substantially no chromatic dispersion.
[0212] In some embodiments, the first diffractive element and the
second diffractive element have substantially identical diffractive
patterns.
[0213] In some embodiments, the second diffractive element is
positioned so a diffractive surface of the second diffractive
element is parallel to a corresponding diffractive surface of the
first diffractive element. In other embodiments, one or more
optical elements (such as dielectric mirrors having very high
reflectivity at the laser wavelengths) are used to reflect the
beams on their path between the diffractive surfaces, and the
second diffractive grating is placed where the reflected beams
converge to a single area, and at an angle selected to remove
further chromatic dispersion.
[0214] In some embodiments, the source of a plurality of light
beams includes a plurality of fiber lasers each tuned to a
different wavelength.
[0215] Some embodiments further include a plurality of Yb-doped
multiply clad optical fibers including a first fiber and a second
fiber each having an inner cladding and one or more outer
claddings, and one or more laser diodes optically coupled to insert
their laser outputs as pump light to the inner claddings of the
plurality of fibers, wherein the first fiber generates a first
laser beam at a first wavelength for one of the plurality of light
beams, and the second fiber generates a second laser beam at a
second wavelength for another of the plurality of light beams.
[0216] Some embodiments further include a tuner apparatus
operatively coupled to the first fiber to set the first laser beam
at the first wavelength, a tuner apparatus operatively coupled to
the second fiber to set the second laser beam at the second
wavelength, a detector operatively coupled to detect whether one of
the laser beams has become misaligned relative to the single
combined beam, a diagnoser operatively coupled to determine whether
the first laser beam is the misaligned one and if so, to adjust the
tuning of the first fiber in order that the first laser beam is
aligned relative to the single combined beam.
[0217] In some embodiments, the detector and the diagnoser base the
determination of whether the first laser beam is the misaligned one
based on a timing of a pulse of one of the laser beams.
[0218] In some embodiments, the detector and the diagnoser are
operable to determine that the first laser beam is the misaligned
one and tune the wavelength of the first laser while the second
beam is active in the single combined beam.
[0219] In some embodiments, the detector and the diagnoser are
operable to determine that the first laser beam is the misaligned
one while a plurality of the other laser beams are ON, and wherein
the diagnoser determines a timing of a change in a power value of
the first laser beam, the detector detects a corresponding change
in a misaligned beam, and the first tuner changes a tuning of the
first laser based on the detection of the corresponding change in
the misaligned beam.
[0220] Some embodiments of the present invention provide an
apparatus that includes means for sourcing a plurality of light
beams, and diffractive means for spectrally combining light from
the plurality of light beams into a single combined beam and, in
some embodiments, for removing some chromatic dispersion from the
single combined beam.
[0221] In some embodiments, the diffractive means includes a
plurality of diffractive surfaces and the light from the plurality
of light-beam sources interacts with the plurality of diffractive
surfaces serially.
[0222] In some embodiments, the diffractive means includes a first
diffractive surface and a second diffractive surface having the
same diffractive pattern.
[0223] In some embodiments, the diffractive means for spectrally
combining includes positioning the second diffractive surface so a
diffractive surface of the second diffractive element is parallel
to the first diffractive surface.
[0224] In some embodiments, the means for sourcing a plurality of
light beams includes means for generating a plurality of laser
beams for use as the plurality of light beams.
[0225] Some embodiments further include a plurality of Yb-doped
multiply clad optical fibers including a first fiber and a second
fiber, means for pumping an inner cladding of the plurality of
fibers with pump light from one or more laser diodes, means for
generating with the first fiber a first laser beam at a first
wavelength for one of the plurality of light beams, and means for
generating with the second fiber a second laser beam at a second
wavelength for another of the plurality of light beams.
[0226] Some embodiments further include means for tuning the first
fiber to generate the first laser beam at the first wavelength,
means for tuning the second fiber to generate the second laser beam
at the second wavelength, means for detecting that one of the laser
beams has become misaligned relative to the single combined beam,
means for determining that the first laser beam is the misaligned
one, and means for adjusting the tuning of the first fiber in order
that the first laser beam is aligned relative to the single
combined beam.
[0227] In some embodiments, the means for detecting of the
misaligned beam and the means for determining that the first laser
beam is the misaligned one are based on a timing of a pulse of one
of the laser beams.
[0228] In some embodiments, the means for detecting of the
misaligned beam and the means for determining that the first laser
beam is the misaligned one perform these functions while the second
beam is active.
[0229] In some embodiments, the means for determining that the
first laser beam is the misaligned one perform this function while
a plurality of the other laser beams are ON, and includes: means
for changing a power value of the first laser beam, means for
detecting a corresponding change in a misaligned beam, and means
for changing a tuning of the first laser based on the detecting of
the corresponding change in the misaligned beam.
[0230] Some embodiments of the present invention provide a method
that includes providing a plurality of laser beams including a
first laser beam and a second laser beam, spectrally combining the
plurality of laser beams into a single output beam, wavelength
tuning the first fiber to generate the first laser beam at the
first wavelength, wavelength tuning the second fiber to generate
the second laser beam at the second wavelength, detecting that one
of the laser beams has become misaligned relative to the single
combined beam, determining that the first laser beam is the
misaligned one, and adjusting the wavelength tuning of the first
fiber in order that the first laser beam is aligned relative to the
single combined beam.
[0231] In some embodiments, the spectrally combining further
includes providing a first diffractive element, introducing a first
chromatic dispersion into the first laser beam with the first
diffractive element, introducing a second chromatic dispersion into
the second laser beam with the first diffractive element, and
providing a second diffractive element, and introducing a third
chromatic dispersion into the first laser beam with the second
diffractive element, wherein the third chromatic dispersion is a
compensating dispersion that negates at least a portion of the
first chromatic dispersion from the first laser beam, and
introducing a fourth chromatic dispersion into the second laser
beam with the second diffractive element, wherein the fourth
chromatic dispersion is a compensating dispersion that negates at
least a portion of the second chromatic dispersion from the second
laser beam.
[0232] In some embodiments, the providing of the first diffractive
element and the second diffractive element includes providing
dielectric diffractive gratings having the same diffractive pattern
on both the first diffractive element and the second diffractive
element.
[0233] Some embodiments further include positioning of the second
diffractive element so a diffractive surface of the second
diffractive element is parallel to a corresponding diffractive
surface of the first diffractive element.
[0234] Some embodiments further include providing a plurality of
Yb-doped multiply clad optical fibers including a first fiber and a
second fiber, pumping an inner cladding of each of the plurality of
fibers with pump light from one or more laser diodes, amplifying,
with the first fiber, the first laser beam at a first wavelength,
and amplifying, with the second fiber, the second laser beam at a
second wavelength.
[0235] Some embodiments further include filtering the first laser
beam to a full-width half-maximum linewidth of about one nanometer
or less, filtering the second laser beam to a full-width
half-maximum linewidth of about one nanometer or less, pulsing the
first laser beam to a pulse length of about ten nanoseconds or
less, and sufficiently short to substantially prevent SBS buildup
in the amplifying of the first laser beam, and pulsing the second
laser beam to a pulse length of about ten nanoseconds or less, and
sufficiently short to substantially prevent SBS buildup in the
amplifying of the second laser beam.
[0236] Some embodiments further include detecting that one of the
laser beams has become angularly misaligned relative to the single
combined beam, determining which laser beam is the angularly
misaligned one, and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam, wherein the detecting of the angularly misaligned beam and
the determining of which laser beam is the angularly misaligned one
are based on a timing of a pulse of one of the laser beams.
[0237] Some embodiments further include detecting that one of the
laser beams has become angularly misaligned relative to the single
combined beam, determining which laser beam is the angularly
misaligned one, and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam.
[0238] In some embodiments, the detecting of the angularly
misaligned beam and the determining of which laser beam is the
angularly misaligned one are performed while one or more of the
other beams are active.
[0239] In some embodiments, wherein the determining of which laser
beam is the angularly misaligned one is performed during a time
when a plurality of the other laser beams are on, and includes
changing a power value of a first laser beam, detecting a
corresponding change in a misaligned beam, and changing the angle
of the first laser based on the detecting of the corresponding
change in the misaligned beam.
[0240] Some embodiments provide an apparatus that includes an
output diffractive element, and a source of a plurality of
substantially monochromatic light beams directed from different
angles to a single location on the output diffractive element,
wherein the output diffractive element spectrally combines the
plurality of light beams into a single beam, and wherein the
plurality of light beams includes a first light beam having a first
central wavelength and a second light beam having a second central
wavelength, a first adjustment apparatus operatively coupled to set
an adjustable characteristic of the first light beam, a second
adjustment apparatus operatively coupled to set an adjustable
characteristic of the second light beam, a detector operatively
coupled to detect whether one of the light beams has become
misaligned relative to the single combined beam, a diagnoser
operatively coupled to determine whether the first light beam is
the misaligned one and if so, to control the first adjustment
apparatus to adjust the adjustable characteristic of the first
light beam in order that the first light beam becomes aligned
relative to the single combined beam.
[0241] In some embodiments, the adjustable characteristic of the
first light beam is the first central wavelength, wherein the
adjustable characteristic of the second light beam is the second
central wavelength, and wherein the first adjustment apparatus
tunes the first central wavelength and the second adjustment
apparatus tunes the second central wavelength.
[0242] Some embodiments further include an input diffractive
element that introduces a chromatic dispersion to each of the
plurality of diffracted light beams, wherein the second diffractive
element introduces a compensating chromatic dispersion to each of
the plurality of diffracted light beams such that the single
combined beam has reduced chromatic dispersion, and wherein the
output diffractive element and the input diffractive element have
substantially identical diffractive patterns.
[0243] In some embodiments, the output diffractive element is
positioned so a diffractive surface of the output diffractive
element is parallel to a corresponding diffractive surface of the
input diffractive element.
[0244] In some embodiments, the source of a plurality of light
beams includes a plurality of MOPA fiber lasers each tuned to a
different wavelength.
[0245] Some embodiments further include a plurality of Yb-doped
multiply clad optical fibers including a first fiber and a second
fiber each having an inner cladding and one or more outer
claddings, and one or more laser diodes optically coupled to insert
their laser outputs as pump light to the inner claddings of the
plurality of fibers, wherein the first light beam is a first laser
beam and the second light beam is a second laser beam, and wherein
the first fiber amplifies the first laser beam at a first
wavelength, and the second fiber amplifies the second laser beam at
a second wavelength.
[0246] Some embodiments further include a first wavelength filter
that filters the first laser beam to a full-width half-maximum
linewidth of about one nanometer or less, a second wavelength
filter that filters the second laser beam to a full-width
half-maximum linewidth of about one nanometer or less, a first
amplitude modulator that pulses the first laser beam to a pulse
length of about ten nanoseconds or less, and sufficiently short to
substantially prevent SBS buildup in the amplifying of the first
laser beam, and a second amplitude modulator that pulses the second
laser beam to a pulse length of about ten nanoseconds or less, and
sufficiently short to substantially prevent SBS buildup in the
amplifying of the second laser beam.
[0247] In some embodiments, the detector and the diagnoser base the
determination of whether the first laser beam is the misaligned one
based on a timing of a pulse of one of the laser beams.
[0248] In some embodiments, the detector and the diagnoser are
operable to determine that the first laser beam is the misaligned
one and tune the wavelength of the first laser while the second
beam is active in the single combined beam.
[0249] In some embodiments, the detector and the diagnoser are
operable to determine that the first laser beam is the misaligned
one while a plurality of the other laser beams are ON, and wherein
the diagnoser determines a timing of a change in a power value of
the first laser beam, the detector detects a corresponding change
in a misaligned beam, and the first tuner changes a tuning of the
first laser based on the detection of the corresponding change in
the misaligned beam.
[0250] Another aspect in some embodiments, provides an apparatus
that includes a source of a plurality of light beams including a
first laser beam and a second laser beam, diffractive means for
spectrally combining the plurality of laser beams into a single
output beam, means for wavelength tuning the first fiber to
generate the first laser beam at the first wavelength, means for
wavelength tuning the second fiber to generate the second laser
beam at the second wavelength, means for detecting that one of the
laser beams has become misaligned relative to the single combined
beam, means for determining that the first laser beam is the
misaligned one, and means for adjusting the means for wavelength
tuning of the first fiber in order that the first laser beam is
aligned relative to the single combined beam.
[0251] In some embodiments, the diffractive means includes a
plurality of diffractive surfaces and the light from the plurality
of light-beam sources interacts with the plurality of diffractive
surfaces serially.
[0252] In some embodiments, the diffractive means includes a first
diffractive surface area and a second diffractive surface area
having the same diffractive pattern.
[0253] In some embodiments, the diffractive means includes means
for positioning the second diffractive surface parallel to the
first diffractive surface.
[0254] In some embodiments, the means for sourcing the plurality of
light beams includes a plurality of Yb-doped multiply clad optical
fibers including a first fiber and a second fiber, means for
pumping an inner cladding of the plurality of fibers with pump
light from one or more laser diodes, means for amplifying with the
first fiber a first seed laser beam at a first wavelength for one
of the plurality of light beams, and means for amplifying with the
second fiber a second seed laser beam at a second wavelength for
another of the plurality of light beams.
[0255] Some embodiments further include means for filtering the
first laser beam to a full-width half-maximum linewidth of about
one nanometer or less, means for filtering the second laser beam to
a full-width half-maximum linewidth of about one nanometer or less,
means for pulsing the first and second laser beams to a pulse
length of about ten nanoseconds or less and sufficiently short to
substantially prevent SBS buildup.
[0256] Some embodiments further include means for compensating for
a chromatic dispersion.
[0257] In some embodiments, the means for detecting of the
misaligned beam and the means for determining that the first laser
beam is the misaligned one perform while the second beam is
active.
[0258] In some embodiments, the determining that the first laser
beam is the misaligned one is performed while a plurality of the
other laser beams are ON, and includes means for changing a power
value of the first laser beam, means for detecting a corresponding
change in a misaligned beam, and means for changing a tuning of the
first laser based on the detecting of the corresponding change in
the misaligned beam.
[0259] In some embodiments, the means for detecting of the
misaligned beam and the means for determining that the first laser
beam is the misaligned one are based on a timing of a pulse of one
of the laser beams.
[0260] Another aspect of some embodiments of the present invention
provides a method that includes providing a plurality of laser
beams including a first laser beam and a second laser beam,
wavelength tuning the first fiber to generate the first laser beam
at the first wavelength and having a linewidth of about 1 nm or
less, wavelength tuning the second fiber to generate the second
laser beam at the second wavelength and having a linewidth of about
1 nm or less, and spectrally combining the plurality of laser beams
into a single output beam using one or more high-efficiency
dielectric diffractive gratings, the output beam having a
power-per-unit-area incident to least one grating of about 10
W/(cm.sup.2 of grating surface) or more.
[0261] In some embodiments, the beam output power/area is about 15
W/(cm.sup.2 of grating surface) or more. In some embodiments, the
beam output power/area is about 20 W/(cm.sup.2 of grating surface)
or more. In some embodiments, the beam output power/area is about
50 W/(cm.sup.2 of grating surface) or more. In some embodiments,
the beam output power/area is about 100 W/(cm.sup.2 of grating
surface) or more. In some embodiments, the beam output power/area
is about 150 W/(cm.sup.2 of grating surface) or more. In some
embodiments, the beam output power/area is about 200 W/(cm.sup.2 of
grating surface) or more. In some embodiments, the beam output
power/area is about 500 W/(cm.sup.2 of grating surface) or more. In
some embodiments, the beam output power/area is about 1,000
W/(cm.sup.2 of grating surface) or more. In some embodiments, the
beam output power/area is about 1,500 W/(cm.sup.2 of grating
surface) or more. In some embodiments, the beam output power/area
is about 2,000 W/(CM.sup.2 of grating surface) or more. In some
embodiments, the beam output power/area is about 5,000 W/(cm.sup.2
of grating surface) or more. In some embodiments , the beam output
power/area is about 10,000 W/(cm.sup.2 of grating surface) or more.
In some embodiments, the beam output power/area is about 15,000
W/(cm.sup.2 of grating surface) or more. In some embodiments, the
beam output power/area is about 20,000 W/(cm.sup.2 of grating
surface) or more. In some embodiments, the beam output power/area
is about 50,000 W/(cm.sup.2 of grating surface) or more. In some
embodiments, the beam output power/area is about 100,000
W/(cm.sup.2 of grating surface) or more.
[0262] In some embodiments, these output power densities are
possible by using high-efficiency dielectric gratings where the
output beam diffracted from the output grating has 80% or more of
the power sum of the input beams, or, in some embodiments, 85% or
more, 90% or more, 95% or more, 97% or more, 98% or more, 99% or
more, 99.5% or more, 99.7% or more, 99.8% or more, or 99.9% or
more, in order to minimize absorption of heat from the diffracted
beam(s). In some embodiments, the gratings are high-efficiency
dielectric reflection gratings, in order to minimize absorption of
heat from the diffracted beam(s). In some embodiments, the gratings
are oriented at or close to the Littrow angle (i.e., where the
output angle .beta..apprxeq..alpha., the input angle, relative to
the grating-surface's normal vector) for at least some of the
wavelengths used, in order to minimize absorption of heat from the
diffracted beam(s).
[0263] In some embodiments, the output beam's power is 500 W or
more. In some embodiments, the output beam's power is 1,000 W or
more. In some embodiments, the output beam's power is 2,000 W or
more. In some embodiments, the output beam's power is 5,000 W or
more. In some embodiments, the output beam's power is 10,000 W or
more. In some embodiments, the output beam's power is 20,000 W or
more. In some embodiments, the output beam's power is 50,000 W or
more. In some embodiments, the output beam's power is 100,000 W or
more. In some embodiments, the output beam's power is 200,000 W or
more. In some embodiments, the output beam's power is 500,000 W or
more. In some embodiments, the output beam's power is 1,000,000 W
or more. In some embodiments, the output beam's power is 2,000,000
W or more. In some embodiments, the output beam's power is
5,000,000 W or more. In some embodiments, the output beam's power
is 10,000,000 W or more.
[0264] In order to obtain high beam quality, M.sup.2, some
embodiments control the FWHM linewidth. In some embodiments, the
first and second (and optionally other subsequent) laser beams each
have a linewidth of about 0.5 nm or less. In some embodiments, the
first and laser beam each have a linewidth of about 0.2 nm or less.
In some embodiments, the first and second laser beam each have a
linewidth of about 0.1 nm or less. In some embodiments, the first
and second laser beam each have a linewidth of about 0.05 nm or
less. In some embodiments, the first and second laser beam each
have a linewidth of about 0.02 nm or less. In some embodiments, the
first and second laser beam each have a linewidth of about 0.01 nm
or less. In some embodiments, the first and second laser beam each
have a linewidth of about 0.005 nm or less. In some embodiments,
the first and second laser beam each have a linewidth of about
0.002 nm or less. In some embodiments, the first and second laser
beam each have a linewidth of about 0.001 nm or less.
[0265] In order to obtain high beam quality, M.sup.2, some
embodiments control spectral fill (i.e., the ratio of FWHM
linewidth/center-to-center wavelength
spacing=.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)) relative to
the spatial fill (i.e., the ratio of beam-waist width
.omega..sub.0/beam center-to-center spacing (X.sub.N+1-X.sub.N)) of
successive input beams. In some embodiments, spectral fill
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N) is set to be equal
to or less than spatial fill .omega..sub.0/(X.sub.N+1-X.sub.N). In
some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<.omega..sub.0/(X.sub.N+-
1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.9.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.8.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.7.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.6.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.5.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.4.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.3.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.2.omega..sub.0/(X.sub-
.N+1-X.sub.N). In some embodiments,
.DELTA..lamda./(.lamda..sub.N+1-.lamda..sub.N)<0.1.omega..sub.0/(X.sub-
.N+1-X.sub.N).
[0266] In order to obtain high beam quality, M.sup.2, some
embodiments reduce the output beam center-to-center spacing
(X.sub.N+1-X.sub.N) by shaving, grinding or otherwise reducing a
diameter of one or more of the output (e.g., final stage of the
power amplifier) fibers. Some such embodiments use a plurality of
cores spaced side-by-side along a straight transverse line of a
"ribbon" fiber, in order to reduce the output beam center-to-center
spacing (X.sub.N+1-X.sub.N). Some embodiments reduce the
center-to-center input spacing (X'.sub.N+1-X'.sub.N) by shaving,
grinding or otherwise reducing a diameter of one or more of the
frequency-setting (e.g., master oscillator) fibers. Some
embodiments use a side or star coupler at or near an output end of
the output (e.g., final stage of the power amplifier) fibers. Some
embodiments use a photonic-crystal fiber, fiber-like rod, or rod as
the output or final stage of the power amplifier.
[0267] In order to obtain high beam quality, M.sup.2, and to also
obtain high power, some embodiments operate a large-mode area
amplifying fiber, or photonic-crystal fiber, fiber-like rod, or
rod, operating substantially on its fundamental mode (i.e., the
LP.sub.01 mode of a fiber, corresponding to a TEM.sub.00 mode of
other-lasers).
[0268] Some embodiments further include temporally forming the
first laser beam into a first serial plurality of pulses and
temporally forming the second laser beam into a second serial
plurality of pulses, each such pulse having a pulse length of about
10 ns or less. In some embodiments, the pulse lengths are about 9
ns or less. In some embodiments, the pulse lengths are about 8 ns
or less. In some embodiments, the pulse lengths are about 7 ns or
less. In some embodiments, the pulse lengths are about 6 ns or
less. In some embodiments, the pulse lengths are about 5 ns or
less. In some embodiments, the pulse lengths are about 5 ns or
less. In some embodiments, the pulse lengths are about 4 ns or
less. In some embodiments, the pulse lengths are about 3 ns or
less. In some embodiments, the pulse lengths are about 2 ns or
less. In some embodiments, the pulse lengths are about 1 ns or
less. In some embodiments, the pulse lengths are about 0.5 ns or
less.
[0269] In some embodiments, pulses of the first serial plurality of
pulses are alternated with pulses of the second serial plurality of
pulses.
[0270] Some embodiments further include detecting that one of the
laser beams has become misaligned relative to the single combined
beam, determining that the first laser beam is the misaligned one,
and adjusting the wavelength tuning of the first fiber in order
that the first laser beam is aligned relative to the single
combined beam.
[0271] In some embodiments, the spectrally combining further
includes: providing a first diffractive element, introducing a
first chromatic dispersion into the first laser beam with the first
diffractive element, providing a second diffractive element,
introducing a second chromatic dispersion into the second laser
beam with the second diffractive element, and providing a third
diffractive element, and spectrally combining the first and second
laser beams and introducing a third chromatic dispersion into the
first laser beam with the third diffractive element, wherein the
third chromatic dispersion is a compensating dispersion that
negates at least a portion of the first chromatic dispersion from
the first laser beam, and introducing a fourth chromatic dispersion
into the second laser beam with the second diffractive element,
wherein the fourth chromatic dispersion is a compensating
dispersion that negates at least a portion of the second chromatic
dispersion from the second laser beam.
[0272] In some embodiments, the providing of the first diffractive
element and the second diffractive element includes providing
dielectric diffractive gratings having the same diffractive pattern
on both the first diffractive element and the second diffractive
element, and positioning the second diffractive element so a
diffractive surface of the second diffractive element is approached
by the laser beams at an angle corresponding to an angle the beams
left the first diffractive element.
[0273] Some embodiments further include providing a plurality of
Yb-doped large-mode-area (LMA) optical-amplification fibers
operating substantially on the fundamental mode, including a first
fiber and a second fiber, pumping of each of the plurality of
fibers with pump light from one or more laser diodes, amplifying,
with the first fiber, the first laser beam at a first wavelength,
and amplifying, with the second fiber, the second laser beam at a
second wavelength.
[0274] Some embodiments further include filtering the first laser
beam to a full-width half-maximum linewidth of about one nanometer
or less; filtering the second laser beam to a full-width
half-maximum linewidth of about one nanometer or less; pulsing the
first laser beam to a pulse length of about ten nanoseconds or
less, and sufficiently short to substantially prevent SBS buildup
in the amplifying of the first laser beam; and pulsing the second
laser beam to a pulse length of about ten nanoseconds or less, and
sufficiently short to substantially prevent SBS buildup in the
amplifying of the second laser beam. For the filtering, in some
embodiments, the full-width half-maximum linewidth is about 0.5 nm
or less. In some embodiments, the full-width half-maximumlinewidth
is about 0.5 nm or less. In some embodiments, the full-width
half-maximum linewidth is about 0.3 nm or less. In some
embodiments, the full-width half-maximum linewidth is about 0.2 nm
or less. In some embodiments, the full-width half-maximum linewidth
is about 0.1 nm or less. In some embodiments, the full-width
half-maximum linewidth is about 0.05 nm or less. In some
embodiments, the full-width half-maximum linewidth is about 0.03 nm
or less. In some embodiments, the full-width half-maximum linewidth
is about 0.02 nm or less. In some embodiments, the full-width
half-maximum linewidth is about 0.01 nm or less. In some
embodiments, the full-width half-maximum linewidth is about 0.005
nm or less. In some embodiments, the full-width half-maximum
linewidth is about 0.003 nm or less. In some embodiments, the
full-width half-maximum linewidth is about 0.002 nm or less. In
some embodiments, the full-width half-maximum linewidth is about
0.001 nm or less. In some embodiments, the full-width half-maximum
linewidth is about 0.0005 nm or less. In some embodiments, the
full-width half-maximum linewidth is about 0.0003 nm or less. In
some embodiments, the full-width half-maximum linewidth is about
0.0002 nm or less. In some embodiments, the full-width half-maximum
linewidth is about 0.0001 nm or less.
[0275] Some embodiments further include detecting that one of the
laser beams has become angularly misaligned relative to the single
combined beam, determining which laser beam is the angularly
misaligned one, and adjusting an angle of the angularly misaligned
laser beam in order to align it relative to the single combined
beam, wherein the detecting of the angularly misaligned beam and
the determining of which laser beam is the angularly misaligned one
are based on a timing of a pulse of one of the laser beams.
[0276] In some embodiments, the detecting of the angularly
misaligned beam and the determining of which laser beam is the
angularly misaligned one are performed while one or more of the
other beams are active. As used herein, two lasers are both ON is
defined to mean instantaneously and simultaneously emitting light,
whether as when two pulse are simultaneously on, or two cw lasers
are both on; whereas two lasers are both ACTIVE is defined to mean
operating in their normal mode, which, when pulsed lasers are
discussed, means where both lasers are pulsing, whether or not the
pulses of one laser are instantaneously and simultaneously emitting
light, as well as when two cw lasers are both on.
[0277] In some embodiments, the determining of which laser beam is
the angularly misaligned one is performed during a time when a
plurality of the other laser beams are on, and includes: changing a
power value of a first laser beam, detecting a corresponding change
in a misaligned beam; and changing the angle of the first laser
based on the detecting of the corresponding change in the
misaligned beam.
[0278] It is specifically contemplated that the present invention
includes embodiments having combinations and subcombinations of the
various embodiments and features that are individually described
herein (i.e., some of the features from one embodiment combined
with some of the features of another embodiment. Further, some
embodiments include fewer than all the components described as part
of any one of the embodiments described herein.
[0279] It is to be understood that the above description is
intended to be illustrative, and not restrictive. Although numerous
characteristics and advantages of various embodiments as described
herein have been set forth in the foregoing description, together
with details of the structure and function of various embodiments,
many other embodiments and changes to details will be apparent to
those of skill in the art upon reviewing the above description. The
scope of the invention should be, therefore, determined with
reference to the appended claims, along with the full scope of
equivalents to which such claims are entitled. In the appended
claims, the terms "including" and "in which" are used as the
plain-English equivalents of the respective terms "comprising" and
"wherein," respectively. Moreover, the terms "first," "second," and
"third," etc., are used merely as labels, and are not intended to
impose numerical requirements on their objects.
* * * * *